# Research Papers — 2026-08 · bundle page 124 of 240 > Full-text markdown of papers #6151–6200 of 11995 published in 2026-08, newest first. AI-readable mirrors of the full 11995-paper 2026-08 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-08/page/124 [← Previous bundle](https://aipod.dev/papers/archive/2026-08/page/123.md) [Next bundle →](https://aipod.dev/papers/archive/2026-08/page/125.md) ### Comparative Analysis of State-of-the-Art Foundation Models for Sleep Analysis Under Channel Reduction Authors: Hassan Mehdi, Riku Klen, Ayse Kosal Bulbul, Suzanne Timmons, Abdulhamit Subasi, Wei Chen, Zou Zhu, Muhammad Irfan Published: 2026-08-14 Categories: eess.SP, cs.LG arXiv: https://arxiv.org/abs/2609.22105v1 Automatic sleep staging from polysomnography (PSG) is a well-studied task, but PSG itself is expensive, clinic-based, and burdensome to manually score, which limits its use for long-term or at-home monitoring. Most existing sleep-staging foundation models are evaluated using the full PSG montage. We instead ask how much of that montage is actually necessary. We evaluate six sleep staging models on the Multi-Ethnic Study of Atherosclerosis (MESA) PSG dataset across three signal conditions: electroencephalography (EEG), electrocardiography (ECG), and their combination (EEG+ECG). This is motivated by edge-cloud deployment, where EEG requires a clinic-grade scalp electrode, whereas ECG is already captured by consumer wearables. We test state-of-the-art foundation models such as SleepFM with an encoder trained from scratch on MESA, alongside BIOT, MOMENT, LaBraM, a base-scale Vision Transformer (ViT-B) reimplementation of SensorLM trained from scratch, and YASA, spanning EEG-pretrained, general-time-series, from-scratch, and classical non-learned approaches. No model architecture is modified from its original form; SensorLM's encoder is reimplemented only in PyTorch. For EEG-only staging, BIOT achieves the best result with a macro~F1 of 0.7237, followed by LaBraM (0.6835) and SleepFM from scratch (0.6582). Across the five models capable of ECG-only staging, switching from EEG to ECG costs between 0.2798 (MOMENT) and 0.4151 (BIOT) macro~F1, averaging 0.3531, while cutting the raw channel data rate to a third. Adding ECG to EEG provides no gain for most models. These results show that EEG carries most of the sleep-staging signal, quantify the consistent accuracy cost of the wearable-compatible alternative, and demonstrate that sleep-relevant pretraining transfers well to MESA. --- ### DeepInstructor: An Agentic AI Instructor for Experience-Driven Idea Evaluation Authors: Rongcan Pei, Fang Guo, Qinglin Qi, Qi Zhu, Yun Luo, Jianhao Yan, Minjun Zhu, Qiujie Xie, Dehong Zheng, Yue Zhang Published: 2026-08-14 Categories: cs.CL, cs.IR arXiv: https://arxiv.org/abs/2609.22104v1 As automated scientific discovery advances, Large Language Models (LLMs) can now generate research ideas at an unprecedented scale, shifting the bottleneck from idea generation to idea evaluation. Existing evaluators mainly rely on parametric LLM knowledge or unstructured retrieval, producing judgments that lack the experience-grounded reasoning used by human instructors. To address this, we propose DeepInstructor, an agentic framework that formulates idea evaluation as reasoning over structured scholarly experience. DeepInstructor constructs an Experience Graph from 58,607 peer reviews and employs a ReAct-based agent to retrieve dimension-specific evidence for traceable evaluation. We further introduce DeepInstruct, a dataset with controlled pairwise comparisons across novelty, significance, and feasibility. Experiments show that DeepInstructor substantially outperforms existing baselines, improving Hit@1 and Hit@2 alignment with human judgments by 24.4% and 29.7%, respectively. Our findings suggest that scientific idea evaluation can be grounded in explicit reasoning over structured scholarly experience --- ### GradRepair-ODE: Certified Gradient Repair for Neural ODE Training Authors: Ziqian Bi, Xin Liang Chia Published: 2026-08-14 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.13204v1 Neural ordinary differential equations use numerical solvers inside the training loop. The solver determines the forward trajectory and also affects the gradient passed to the optimizer. That coupling creates a reliability problem for scientific machine learning and continuous-time generative modeling, including diffusion probability-flow ordinary differential equations and flow-matching models. Under loose step sizes, stiff dynamics, chaotic sensitivity, or event discontinuities, a differentiable ODE pipeline can return a finite gradient whose direction is numerically suspect. We introduce GradRepair-ODE, a reliability framework for checking, repairing, and rejecting ODE gradients at the optimizer step. The method computes several gradient candidates, compares them with directional finite-difference checks and solver diagnostics, diagnoses likely numerical failure modes, repairs selected gradients through path switching or stricter recomputation, and rejects steps whose descent direction cannot be certified. In six synthetic ODE systems, GradRepair-ODE leaves low-risk systems unchanged, repairs Robertson and Lorenz gradients to cosine similarity 1.000 against a strict reference, reduces unsafe accepted steps from 37 to 0, and rejects an event-discontinuous case instead of applying an uncertified update. The paper argues for a simple change in the training contract: an ODE gradient should reach the optimizer with numerical evidence attached. --- ### Scalable partial information decomposition for symptom networks via supervised embeddings Authors: Cillian Hourican, Eric Dignum, Rick Quax, Debraj Roy Published: 2026-08-14 Categories: cs.LG, cs.IT arXiv: https://arxiv.org/abs/2609.13203v1 Pairwise relationships among mental-health symptoms are routinely summarised asscalar edge weights, which cannot express whether two symptoms carry overlapping information about a third or information that appears only in combination. Partial information decomposition (PID) addresses this gap but is computationally intractable beyond a few sources. We introduce embedding-based PID (ePID), a scalable pipeline that compresses all nonfocal symptoms into a low-cardinality discrete embedding and computes a tractable two-source PID, yielding source-unique, remainder-unique, redundant, and synergistic components for each ordered source-target pair. We benchmarked 13 candidate embeddings on synthetic Bayesian networks calibrated to PHQ-9 and on 83 real-world datasets across five PID measures. A supervised Agglomerative Conditional Information Bottleneck (ACIB) embedding recovered the reference decomposition most accurately of the 13 embeddings tested, and did so for every PID measure yielding non-negative atoms once four or more symptoms were compressed (synergy recovery r = 0.92). The two instruments then diverged sharply. In PHQ-9 networks (UK Biobank,N = 154,291; Xinxiang student sample, N = 24,292) the surrounding symptom context carried most pairwise dependence through redundant and remainder-unique channels; synergy contributed 6 to 9%, and no directed edge was synergy-dominated in either cohort. In the 28-item Interpersonal Reactivity Index, 45% of source pairs were. The identical pipeline, applied without parameter changes, therefore returned opposite profiles for the two instruments, each consistent with how that instrument was constructed. By separating overlapping from interaction-dependent information, ePID provides a scalable, model-agnostic complement to standard symptom-network methodology to distinguish redundant and synergistic contributions to observed correlations. --- ### Do Tabular Foundation Models Still Need Feature Engineering? Authors: Yifan WU, Pinjun Dong, Jiran Tao, Binyan Jiang Published: 2026-08-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.13202v1 Feature engineering has long been a cornerstone of tabular machine learning. Tabular foundation models (TFMs) are pretrained on a wide range of tabular datasets and applied via in-context learning. Their rise raises a natural question: does manual feature construction still matter as these models become more capable? To answer this, we perform a controlled study across several versions of two major TFM families, testing a wide range of existing feature engineering techniques on benchmark datasets from TabArena. We find a consistent pattern: feature engineering gains are concentrated in earlier model generations and become negligible for the strongest models. These results suggest that stronger TFMs depend less on explicitly engineered input representations. In a complementary experiment, however, adding in-context information from related datasets still improves performance. Our findings indicate a shift in the source of performance gains for stronger TFMs: re-representing existing inputs becomes less effective, while providing additional task-relevant context remains beneficial. --- ### Criticality in Dissimilar Decomposition and Undersampling of Random Datasets with Anomalies Authors: Ghurumuruhan Ganesan Published: 2026-08-14 Categories: cs.LG, cs.IT, math.PR arXiv: https://arxiv.org/abs/2609.13201v1 Training datasets for upcoming LLMs would include a significant amount of AI text/image data generated from current LLMs. In such a scenario, it is important to understand how this affects batch decompositions and thereby, the performance of the resultant new LLM. In this paper, we consider AI generated data as anomalies ``linked" to main data points and study decomposition and undersampling properties of the overall random dataset. We use redundancy graphs and iteration techniques to obtain bounds for the minimum size of a strongly dissimilar (SD) decomposition and demonstrate a phase transition phenomena, wherein the minimum size is essentially determined by the \emph{main} data points when the number of anomalies is small and is ``taken" over by the anomalies above a certain threshold. We also establish a size criticality result for the strong similarity of a randomly undersampled dataset and illustrate our results with examples involving categorical datasets, whose overall space size is much larger than the size of the dataset. --- ### PRISM-UDE: Physics-Regularized Iterative Symbolic Modeling of 3nm FinFETs via Universal Differential Equation Authors: Pranavanath Balamurali, Prathamesh Dinesh Joshi, Raj Abhijit Dandekar, Rajat Dandekar, Sreedath Panat Published: 2026-08-14 Categories: cs.ET, cs.LG arXiv: https://arxiv.org/abs/2609.13200v1 Compact transistor models are the mathematical backbone of circuit simulation. However, at advanced nodes such as 3nm, transport physics becomes too complex for traditional hand-derived equations to capture accurately. Purely data-driven neural surrogates, on the other hand, are numerically unstable inside circuit solvers and offer no physical insight into their own predictions. We introduce PRISM-UDE (Physics-Regularized Iterative Symbolic Modeling via Universal Differential Equations), a framework that embeds a small neural network inside a physics-based transistor model, using the network only to learn the transport behavior that the analytical baseline misses, rather than replacing the physics altogether. Once trained, this neural correction is distilled into a single, interpretable closed-form expression via symbolic regression, making the final model fully analytical and simulator-ready. Applied to a 3nm FinFET benchmark dataset, PRISM-UDE reduces prediction error more than sixfold (70.33% to 11.01%) relative to the standard physics-only baseline. The distilled expression preserves this accuracy almost exactly while eliminating the neural network entirely. We further validate the extracted expression directly inside a SPICE circuit simulator, confirming stable, physically consistent behavior under both static bias sweeps and dynamic switching conditions. --- ### Discovering and Preserving Category Correlation Knowledge via Adaptive Reciprocal Knowledge Distillation Authors: Dawen Jiang, Zhishu Shen, Zeyu Liu, Tiehua Zhang Published: 2026-08-14 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2609.13199v1 Knowledge distillation aims to improve the performance of lightweight student models by transferring knowledge from larger and more powerful teacher models. However, a substantial size gap between teacher and student models often impedes effective knowledge transfer. Most existing approaches adopt a static, one-way teacher-to-student distillation paradigm, which overlooks the dynamic nature of student learning and fails to provide targeted guidance on hard samples. In this paper, we propose adaptive reciprocal knowledge distillation (AR-KD), a novel method that improves knowledge transfer by simplifying the teacher's output distribution. Specifically, AR-KD performs reciprocal adaptation on the teacher by matching its class correlation matrix to the student's relational representation, which reshapes the teacher's prediction structure to better suit the student's capacity. This relational alignment mitigates the collapse of inter-class dark knowledge caused by overconfident teachers, enabling the student to learn from richer and more compatible supervisory signals. We evaluate AR-KD on CIFAR-100 and ImageNet-1k classification datasets, where it outperforms state-of-the-art knowledge distillation baselines. Specifically, AR-KD improves student performance across homogeneous and heterogeneous setups: up to 7.13% accuracy gain for students, 1.42% to 4.15% higher than vanilla KD on average, and further improvements when integrated with other advanced methods. Our code is available at https://anonymous.4open.science/r/ARKD/. --- ### SPICE: Simple Polysemantic Feature Interpretation via Clustering-based Explanation Authors: Sehyun Lee, Dahee Kwon, Damin Lee, Jaesik Choi Published: 2026-08-14 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2609.13198v1 One of the pivotal recent challenges in neural network interpretability is polysemanticity, where a single neuron is activated by multiple, often unrelated concepts, hindering clear functional understanding. Although prior work has explored this phenomenon, existing approaches remain architecture-specific and depend on manual heuristics such as a fixed number of concept clusters ($K$), limiting their generality and scalability--especially for modern Transformer-based models. To address these limitations, we introduce SPICE (\textbf{S}imple \textbf{P}olysemantic Feature \textbf{I}nterpretation via \textbf{C}lustering-based \textbf{E}xplanation), a generalizable framework for analyzing polysemanticity in deep vision architectures. SPICE avoids architecture-dependent propagation rules, enabling the first systematic comparison of polysemanticity across both CNNs and Transformers, and automatically determines the number of concept clusters per neuron, eliminating reliance on a preset $K$ and supporting scalable analysis for large models. Using SPICE, we conduct a comprehensive investigation into how polysemanticity emerges, varies across depth and architecture, and forms through distinct computational pathways. --- ### Algorithmic Information Dynamics of Learning: A Certified, Differentiable Complexity Controller for Grokking Authors: Luan Ozelim, Abicumaran Uthamacumaran, Hector Zenil Published: 2026-08-14 Categories: cs.LG, cs.IT arXiv: https://arxiv.org/abs/2609.13197v2 Algorithmic Information Dynamics (AID) studies systems by perturbing them and measuring changes in algorithmic complexity, but its usual estimator, the Block Decomposition Method, is piecewise constant, restricting the calculus to finite differences. We use $K^{\mathrm{CDM}}_{\mathrm{s}F}$, a certified, differentiable estimator, to bring the calculus into learning dynamics: grokking, where a complexity order parameter is known but has not been made to act. As a transient loss kick, the estimator becomes a controller that accelerates grokking in Levin's description-length--versus-time sense, within a data-dependent Occam boundary whose finite-size trend, $f_c\sim\ln p/p$, is consistent with a coupon-collector interpretation. Ablations show that a complexity gate matches a train-loss gate in rescuing failing seeds with $27\%$ less intervention; among the tested signals, only map complexity marks the transition's completion; the certified prior and the per-parameter $\nabla K$ attribution are both fungible (a uniform-prior sensor makes bit-identical gate decisions, and random supports match $\nabla K$-selected ones above a sparsity threshold); and direct field perturbation shows a nucleation-like response to the Occam field (no linear regime is resolved over the probed amplitudes, so these measurements do not justify a fluctuation--dissipation surrogate), with a finite-field response growing by orders of magnitude toward the phase-transition. These measurements account for the empirically tuned staircase: bang--bang pulses, stall-fired and released on yield, whose iteration plausibly builds the response it exploits. The kick transfers to sparse parity and to a transformer; a sustained weight-space loss fails. The algorithmic estimator's distinct contribution is timing (when to fire and when to release), not attribution. --- ### OpenAgentFlow: Enabling System-Wide Safety Boundaries for Heterogeneous AI Agent Fleets Authors: Dongsheng Chen, Xiangyu Zhao, Xin Yao, Xuetao Wei Published: 2026-08-14 Categories: cs.AI, cs.CR arXiv: https://arxiv.org/abs/2609.00015v2 AI agents powered by large language models are evolving from isolated assistants into heterogeneous systems in which multiple agents, planners, tools, and execution backends operate over shared environments. In such settings, safety becomes a system-level action-governance problem: deciding whether a pending action should be committed given policy-relevant state accumulated across a session. Existing safeguards operate at fragmented boundaries, making it difficult to enforce shared policies over composed action flows across heterogeneous execution paths. We present OpenAgentFlow, a control-plane/action-plane architecture that establishes the action-commit boundary as a shared enforcement interface. GUI, API, tool, and LLM-generated actions are normalized into a common AgentEvent stream and mediated by a shared pre-execution Policy Enforcement Point, while provenance, session state, audit evidence, and updatable policies are maintained outside individual agents. This provides a common governance layer across incompatible executors and allows new policies to take effect without modifying agents, prompts, models, or execution paths. We evaluate OpenAgentFlow through complementary system evaluations spanning controlled action-flow tests, a public external benchmark, policy updates, and real Android execution. On a 300-case controlled suite, OpenAgentFlow achieves 94.00% accuracy and a 95.35% attack-block rate. On the complete 1,220-case AgentDojo-Traj split of TS-Bench, it achieves 97.62% accuracy, 96.59% unsafe-action recall, and a 1.96% safe false-intervention rate. New control-plane rules take effect without modifying protected agents, and the same enforcement path operates across live GUI, API/tool, and LLM-planned Android execution. These results show that a shared action-commit boundary provides a practical basis for system-wide governance across heterogeneous agent execution paths. --- ### BiasMix-Finance: Post-Generation KYC Guardrails for LLM Portfolio Advice Authors: Gaurav Kukreja, Parul Kukreja, Mohammed Abraar, Raj Dandekar, Rajat Dandekar, Sreedath Panat Published: 2026-08-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.28646v1 Large language models (LLMs) can generate plausible-sounding ETF portfolios while silently violating basic KYC-style constraints on risk, fees, and diversification. This is especially problematic in agentic multi-turn advisory systems, where each draft recommendation can become an action unless guarded by an auditable enforcement layer. We study a model-agnostic, asset-agnostic post-generation guardrail pipeline: (i) enforce a strict JSON allocation schema, (ii) validate allocations against numeric caps, and (iii) when violations occur, deterministically project the output to the nearest feasible portfolio via a convex quadratic program (QCQP). We introduce BiasMix-Finance (Mini), a compact stress-test benchmark for constrained decision-making under biased LLM generations, with a 16-ETF universe, three investor profiles, and eight bias prompts. Across three models and three inference modes (direct, critique, self-consistency), first-pass generations violate at least one cap in 47.6-85.7% of test cases (67.2% pooled), but the convex projection layer reduces final feasibility violations to 0% while requiring only a small correction distance (test pooled median D=||w*-w0||_2=0.066), indicating that the guardrail typically preserves the intent of the original allocation. We report violation rates and correction distances with confidence intervals, and paired model comparisons with multiple-testing correction. To support reproducibility, we release the dataset, prompts, caps, and code in our public GitHub repository. --- ### post-graph-rag: A PostgreSQL-Native Bi-Temporal Graph RAG Engine with Temporal Grounding at Synthesis Authors: Chandan Rajah Published: 2026-08-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.24921v3 Graph RAG connects facts no single passage states, but implementations pay three times: in infrastructure, keeping vector store, graph database and document store in sync; in quality, because a pipeline that never refuses extractor output stores edges asserting nothing; and over time, because a graph that only accumulates treats superseded and current facts alike. post-graph-rag is an open-source engine addressing all three. Chunks with embeddings, a canonical entity graph and community summaries live in one PostgreSQL database, with pgvector for search and edge tables for traversal. Extraction output is validated before writing: vague predicates, pronominal names and bare quantities are rejected, predicates normalise onto an optional vocabulary, and entities resolve to one vertex per canonical name. A bi-temporal layer records when a relation held and when the system believed it, superseding incompatible earlier assertions from document order. Against LightRAG on three corpora with extraction and embedding models fixed, it builds a denser and more queryable graph, and supersedes relationships where a baseline with no temporal model supersedes none. On LongMemEval, 500 questions of long-horizon chat memory, it scores 94.0 percent with gemini-3.6-flash against 71.2 for Zep's gpt-4o and 60.2 for full context, leading on all six question types. The largest single contribution is temporal grounding in the prompt: carrying each relation's validity period through to synthesis moves temporal reasoning from 0.496 to 0.881, ablated paired on one graph per instance. On ECT-QA, earnings-call transcripts restating every metric each quarter, it scores 0.807 under that benchmark's own protocol against 0.599, 0.406 and 0.405 published for TG-RAG, LightRAG and GraphRAG. Code: post-graph-rag https://github.com/crajah/post-graph-rag; post-graph https://github.com/crajah/post-graph --- ### Topology of a Smile: Persistent Homology in Dental Imaging Authors: Leon Dahlmeier, Sara Kališnik, Albert Mehl, Bastian Rieck Published: 2026-08-14 Categories: cs.CV, math.AT arXiv: https://arxiv.org/abs/2608.21422v1 CBCT (Cone Beam Computed Tomography) scans provide detailed three-dimensional images, widely used in dentistry for diagnostic and treatment planning tasks. While invaluable, analyzing and documenting these scans is labor-intensive, prompting efforts to automate key steps like the classification and segmentation of anatomical structures to identify tooth types and associated pathologies. In this article, we propose an approach to automation that leverages persistent homology, a framework from topological data analysis that studies the shape of data by identifying features like connected components, holes, and voids across multiple scales. Persistent homology, together with a support vector machine, allows us to classify teeth in a CBCT scan and to perform diagnostics. Our method advances the state of the art, reaching average accuracy scores of 97.67% for tooth-labeling and 96.77% for diagnostic tasks, outperforming a CNN trained on the same data with accuracy of 70.27% and 86.67%, respectively. --- ### MicroPython and CircuitPython: Pythons Quiet Takeover of IoT and Robotics Authors: Sayed Mahbub Hasan Amiri, Atiar Zahan Published: 2026-08-14 Categories: cs.PL, cs.CL, cs.SE arXiv: https://arxiv.org/abs/2608.18160v1 Background: Python has become the dominant language in software and data science, yet embedded systems have remained tied to C/C++ due to performance and memory constraints. MicroPython and CircuitPython are changing this by bringing Python to microcontrollers, lowering barriers for IoT and robotics development. Aim: This article examines whether these platforms are achieving a quiet takeover of embedded systems, focusing on ecosystem growth, practical applications, performance trade-offs, educational adoption, and prospects. Methods: A mixed-methods design was used, including quantitative analysis of GitHub, Stack Overflow, and Google Trends data; curation of case studies from Hackster.io, Hackaday.io, and the Adafruit Learning System; and original benchmarks on ESP32 and Raspberry Pi Pico comparing MicroPython, CircuitPython, and Arduino C++ across GPIO, I2C, SPI, Wi-Fi, and memory usage. Results: Metrics show sustained growth, with MicroPython supporting over 200 boards and CircuitPython over 400. Benchmarks reveal 10-20 times slower I/O and four to six times higher memory use than C, but performance remains adequate for common sensor and network tasks. Case studies demonstrate successful deployment in home automation, robotics, wearables, agriculture, and professional prototyping. Education emerges as a primary adoption driver. Conclusions: Python is not replacing C/C++; rather, it is becoming the default prototyping and educational language for embedded systems. Continued hardware improvements, better tooling, and standardization will likely deepen this trend. The article offers balanced, evidence-based insights for developers, educators, and technology decision-makers. --- ### Trust Is Not Enough: Influence Calibration for On-Policy Self-Distillation in Agentic RL Authors: Qizhen Lan, Xi Xiao, Xiangchen Guan, Mengchen Fan, Moule Lin, Jung Im Choi, Lijing Zhu Published: 2026-08-14 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.14945v1 On-policy self-distillation (OPSD) gives language agents dense token-level supervision from a privileged self-teacher on the policy's own trajectories. Existing methods allocate this supervision mainly by teacher trust, but trust does not reveal whether emphasizing a token supports the current policy objective. We call this the trust-utility mismatch and introduce Influence Calibration for Self-Distillation (ICSD). For each supervised token, ICSD measures the first-order response of its importance-weighted RL surrogate contribution to a teacher-directed output perturbation. Batch-adaptive calibration converts this non-stationary signal into a bounded allocation weight while preserving the original auxiliary-loss mass within each action turn. These detached weights affect only the distillation loss and require no additional model pass. Across ALFWorld, WebShop, and Search-QA, ICSD improves all matched aggregate metrics over trust-only allocation under Group Relative Policy Optimization (GRPO) and Group-in-Group Policy Optimization (GiGPO), across two model families spanning 1.5B to 7B. At 7B, it reaches 96.1% ALFWorld success and a WebShop score of 93.1. Frozen-batch analyses show that ICSD reduces teacher-supported mass assigned to objective-opposed tokens from 60.1% to 37.8% and raises cosine compatibility with the RL gradient by 0.192. A companion repository is avail- able at https://github.com/lanqz7766/Influence-Calibration-for-On-Policy-Self-Distillation-in-Agentic-RL. --- ### SkillComposer: Learning Reusable Skills for Natural-Language Robot Programming Authors: John Woods, Hasti Seifi Published: 2026-08-14 Categories: cs.RO, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.14944v1 Natural-language interfaces can lower the barrier to programming robots, but existing systems struggle when users request complex tasks. While large language models (LLMs) perform well with simple commands, they often struggle to generate code for multi-step tasks, decompose high-level instructions, or reuse prior solutions. We present SkillComposer, an interactive natural-language robot programming system for simulation environments that continually learns reusable program abstractions. SkillComposer uses a generate-test architecture in which an LLM iteratively generates and revises robot programs before execution. Successful programs are stored and processed by an online library-learning algorithm that compresses recurring function sequences into reusable macro skills for future tasks. We evaluate SkillComposer through ablation experiments and a user study with 12 participants to determine its effectiveness on manipulation and robot caregiving tasks. The results show that evaluator-guided generation and learned abstractions improve success rates and usability while reducing user effort in natural-language robot programming. --- ### Skill Blocks: How Should an Agent Load Its Skill? A Caching-Correct Comparison of Pre-load, On-Demand Tool-Loading, Progressive Disclosure, and Hybrid Authors: Hironobu Nakasuji Published: 2026-08-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.14943v1 Agent skills are often injected in full on every request, increasing token cost. We compare four content-preserving loading methods: Full, Skill Block, Reference, and Hybrid. Across SearchQA, SpreadsheetBench, ALFWorld, ScienceWorld, and SynthProc, we measure token usage using raw input for single-turn tasks and cache-correct effective input for multi-turn tasks. Results show no universal winner. Hybrid reduces input by 27.4% on SearchQA and 39.8% on SpreadsheetBench. On large multi-turn skills, Skill Block and Hybrid achieve substantial reductions, reaching 62.5% and 52.8% on ScienceWorld and 73.0% and 66.6% on SynthProc. ALFWorld shows smaller gains because procedures are short and repeatedly needed. Paired outcome tests detect no quality differences, though they do not establish equivalence. Overall, conditional loading is most beneficial when large portions of a skill are not needed on every turn. --- ### Looks Can be Deceiving: Annotator and Reviewer Performance Across Imagery Sources in Crowd-Sourced Aerial Damage Assessment Authors: Thomas Manzini, Priyankari Perali, Raisa Karnik, Stephen Johnson, Robin R. Murphy Published: 2026-08-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.14942v1 This paper presents the first known empirical investigation of annotator and reviewer performance across multi-source remotely sensed imagery, evaluating human labeling across drone, crewed aviation, and satellite views. Because existing aerial imagery datasets rely predominantly on single-source imagery, there is no currently established state of practice for efficiently allocating human labor to curate large-scale, multi-source aerial datasets. This work addresses this limitation by analyzing annotator and reviewer performance within a post-disaster building damage assessment dataset of 9 disasters, where 20041 buildings in drone, 20695 buildings in crewed aviation, and 33392 buildings in satellite imagery were labeled. These labels, provided by 187 annotators, were then refined through two successive quality-control stages: a single-reviewer pass followed by a consensus-committee review. Our analysis reveals two findings that raise questions for standard crowd-sourcing practices. First, initial annotations were revised by the final committee at rates that rise steeply from higher- to lower-resolution sources (25.27% for crewed aviation and 36.95% for satellite), with the same ordering at every observed workflow stage. Second, a single individual review reduced but did not resolve this disagreement: after review, the committee still revised 6.85% of drone, 14.05% of crewed, and 20.86% of satellite labels. These observations suggest that, in workflows like this one, uniform review allocation leaves the most residual disagreement in lower-resolution imagery. Based on this evidence, and consistent with prior work on adaptive task assignment and budget-aware quality control, this paper offers three recommendations for multi-source dataset curation. --- ### Degeneracy Counting Quantum Algorithm using Decoherence Authors: Malay Marut Das, Mark A. Novotny, Yaroslav Koshka Published: 2026-08-14 Categories: cs.LG, quant-ph arXiv: https://arxiv.org/abs/2608.14941v1 Counting the global optima of a classical optimization problem is a #P-hard task. We develop the canonical thermal pure quantum (CTPQ) state-based degeneracy counting (CTPQsd#) algorithm that determines the number of global optima of a classical optimization problem P by measuring only a small probe S, without finding individual minima. The method exploits a perturbative relation between the decoherence measure of S and the degeneracy of P when S and P are together in a CTPQ state. We provide the first numerical demonstration that this relation can be used to count the global minima, applying it to problems encoded by diagonal random-energy Hamiltonians as a maximally unstructured testbed for classical binary optimization problems. Classical simulations of up to 20 problem qubits quantify the algorithm's sensitivity to variations in the temperature of the CTPQ state, the Hamiltonian energy range, the problem size, and degeneracy. We establish the temperature threshold for determining the exact degeneracy and identify a second, lower threshold that provides a temperature window to count near-degenerate minima within a user-defined energy tolerance. By confining measurement to S, the protocol replaces tomography over the exponentially large problem Hilbert space with tomography over a small probe represented by only four qubits. --- ### When Is an Agent Evaluation Over? Outcome Finality and Cross-Unit Separation Authors: Avyay M. Casheekar, Hariganesh Tangirala Published: 2026-08-14 Categories: cs.AI, cs.CY arXiv: https://arxiv.org/abs/2608.14940v3 Agent evaluations commonly score the state observed when a run stops and count the run as one trial. Interpreting that score as a final result from a separate trial requires outcome finality and cross-unit separation. Outcome finality requires that later events cannot change the claimed result, while cross-unit separation requires that earlier runs cannot change the relevant conditions of later ones. The endpoint establishes neither condition by itself, and the two can hold independently. Waiting for a delayed outcome may settle the label even though its state remains available to another run. Isolation may prevent carryover even though the scored outcome remains unresolved. We develop a completion argument that identifies the evidence needed for each decision. A final success or failure label is justified only when every relevant effect is resolved or bounded tightly enough to fix the outcome. Any remaining uncertainty must be reported. First, in a controlled replay with fixed agent actions, we find that endpoint and terminal labels differ for every nonzero-delay operation and that a delayed write changes the next run's score under shared state but has no such effect after namespacing or verified reset. Second, in a review of ten public protocols, we find that reset or deliberate retention is documented explicitly more often than unfinished operations or evidence for separate scoring. Finally, we propose an open-effects record for operations and resources that may remain relevant after the endpoint, their status, and their possible effects on the scored outcome or another run. --- ### Small Models Scout Bottleneck Order for Large-Model Data Control Authors: Seungmin Choi, Jiwon Sung, Muhammad Umer, Abhiram Rao Gorle, Guijin Son, Youngjae Yu, John M. Cioffi Published: 2026-08-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.14936v1 Small proxy models are commonly used to identify data mixtures for larger-scale training. We ask whether their training trajectories reveal another transferable structure: the order in which larger models should resolve skill bottlenecks. We formulate first-passage skill training, where each monitored skill has a target floor and the objective is to minimize the tokens required to reach all floors. We introduce LogFloor, a closed-loop controller that directs each round toward current bottlenecks, producing phase-ordered resolution trajectories. Across five bAbI skill slices on Qwen2.5-1.5B, LogFloor reduces token cost by 56.2% on average. In 70M-to-12B transfer, three-round replay of a 70M scout path reaches every floor in all eight target runs, saving 30.9% by pair mean, 39.4% in pooled training tokens, and 37.6% under source-cost accounting. On MMLU-control, a frozen scout path succeeds across all eight 12B runs. Collapsing a path to its static marginal mixture or reversing its phase order removes most benefits, while bottleneck labels alone remain partially useful. These results identify phase-ordered bottleneck resolution as a transferable curriculum structure for monitored skill-targeted training. --- ### Developing an Offshore Machine Learning Surface Layer Scheme Authors: Susan Dettling, Sue Ellen Haupt, Thomas Brummet, Patrick Hawbecker, Branko Kosović, David John Gagne Published: 2026-08-14 Categories: physics.ao-ph, cs.LG arXiv: https://arxiv.org/abs/2608.14935v1 Turbulent fluxes between the surface and the atmosphere are typically parameterized using empirically fit relationships. Here we test machine learning techniques for fitting the relationship for the offshore environment. To do that, data from three offshore sites are used: the Martha's Vineyard Coastal Observatory (MVCO) air-sea interaction tower, the FINO1 research platform, and the CASPER-West FLIP research vessel deployed off the coast of California. Two machine learning methods were employed: Neural Networks (NN) and Random Forests (RF). Because the observational sites had towers with measurements at different levels, the vertical differences were input as gradients. Models were built for both momentum flux and heat flux. ML models trained at the individual sites were competitive with and in some cases, better than the physically-based COARE-3 model tailored to offshore fluxes. The heat flux ML models generally outperformed the physics-based parameterizations for most metrics, but the results were mixed for momentum flux, with only the site with the most training data (MVCO) producing results better than COARE-3. When the ML models from that site were applied to the other sites, results were degraded from using data from the site being tested. ML models built from data combined from the three sites generally showed improvements for the sites with less available training data. When assessing which variables were most important, the wind speed was most important for momentum flux and temperature gradient for heat flux. --- ### Training Leaves Traces: Centered Residual Signatures for Language Model Lineage Verification Authors: Aman Singh Thakur, Rayan Khoury Published: 2026-08-14 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.14929v2 Open-weight language models are fine-tuned, quantized, pruned, and merged, yet their provenance is often undocumented. We study data-free white-box lineage verification: can weights alone reveal whether two compatible model checkpoints share ancestry? Residual training produces a shared identity-aligned component in branch products, so this structure alone cannot establish ancestry. We remove it and compare checkpoint-specific structure across residual blocks, yielding a symmetric lineage score calibrated against independent checkpoints. On residual-MLP and GPT-2 benchmarks, the score separates fine-tuned, LoRA-merged, pruned, and quantized descendants from independent and distilled models (AUROC=1.0), distinguishing weight ancestry from behavioral similarity. Under function-preserving checkpoint laundering experiments, weight-space baselines lose margin or fail; our score remains unchanged and runs 76x faster than the nearest robust baseline on GPT-2. The projection-pairing signal appears across six language-model families and beyond, and a case study correctly identifies 3 related and 7 unrelated LLaMA-2 public checkpoints. Collectively, these results establish a passive, data-free provenance signal for compatible open-weight language-model checkpoints --- ### LLMs Can Predict Failure Risk, But Struggle to Predict Which Collaboration Protocol Pays Off: Cost-Aware Protocol Routing Across Reasoning Tasks Authors: Chih-Hsuan Yang, Jingyan Jiang, Cheng-Hau Yang, Vikram Vasudevan, Huihuo Zheng, Venkatram Vishwanath, Rajeev Thakur Published: 2026-08-14 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.14927v1 Multi-agent large language model (LLM) systems can improve reasoning by spending more computation, but deployment requires deciding when extra collaboration is worth its cost. We isolate this decision by running every problem under four protocols while holding the solver fixed within each setting: direct solving (Baseline), iterative self-correction (Single), planner-executor-reviewer collaboration (PER), and multi-agent deliberation (Broadcast). The primary benchmark comprises 4,181 competition-level math problems; paired robustness checks cover four benchmarks spanning competition math, biology, and broader science with two solver families. Across fixed policies, trained routers, and frozen LLM routers, conservative policies under-escalate, whereas higher-solve frozen routers often over-escalate. A post-answer, pre-collaboration gpt-oss-120b probe ranks Baseline failures with 0.8847 AUROC (4,151 parseable cases; 95% CI (0.8732, 0.8955)). The same score remains informative for predicting whether any collaboration helps (0.7683 AUPRC), but is much weaker for identifying PER- or Broadcast-specific value (0.1674 and 0.1041 AUPRC). Separately, the pre-answer self-confidence gate reaches 78.0% solve at 45K tokens, compared with 73.8% at 71.3K for a frozen gpt-oss-120b router and 92.4% for a retrospective fixed-order oracle. Across 10 paired model-condition settings, the oracle adds 23.2-58.3 points of retrospective coverage over Baseline, but protocol profiles vary by task. In the six settings with held-out router evaluations, oracle gaps remain 18.5-28.9 points. Confidence can therefore support initial escalation, while protocol-specific cost-aware routing remains unresolved. --- ### PaSTel: Anchoring Histology in Spatial Transcriptomics via Multi-Scale Hierarchical Bio-Prior Contrastive Pretraining Authors: Azim Dehghani Amirabad, Junchao Zhu, Pushpak Pati, Walid Abdelmoula, Tommaso Mansi, Rui Liao Published: 2026-08-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.14924v1 Spatial transcriptomics (ST) links tissue morphology with molecular programs, motivating multimodal pretraining methods that align histology images with gene expression. However, existing approaches suffer from two key limitations: spatially informative gene selection is often dominated by ubiquitous housekeeping genes, leading to weakly discriminative representations, and independent spot-patch alignment fails to capture spatial dependencies that are critical for tissue organization. To address these challenges, we introduce PaSTel, a hierarchical multimodal pretraining framework that integrates biological priors at three levels. At the spot level, TF-IDF reweighting is used to identify spatially informative genes; at the functional level, curated KEGG pathways serve as anchors for encoding global biological semantics; and at the regional level, spatial clustering aggregates neighboring spots to model meso-scale tissue structure. Across multiple downstream tasks, PaSTel consistently outperforms existing vision and vision-omics encoders, demonstrating that incorporating multiscale biological priors yields more informative and transferable representations for spatial transcriptomics. --- ### SpIn-ViT: Designing a Sparsity-Induced Vision Transformer That Is Mechanistically Interpretable Authors: Philip H. Lee, Parth Padalkar Published: 2026-08-14 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.14922v1 Mechanistic interpretability has recently expanded to Vision Transformers (ViTs), with Sparse Autoencoders (SAEs) increasingly used as post-hoc tools to decompose internal representations into sparse and more interpretable features. However, because post-hoc SAEs are trained on frozen representations after the ViT has already been optimized, their latent features are not directly aligned with the downstream classification objective. We introduce SpIn-ViT, a framework that jointly trains a pretrained ViT and a modified SAE end-to-end, directly aligning sparse patch-level representations with image classification. SpIn-ViT learns semantically coherent neuron activations that localize meaningful image regions while maintaining competitive predictive performance. We evaluate SpIn-ViT across nine image-classification benchmarks using classification accuracy, quantitative interpretability metrics, AI-based and Human evaluations. Compared with the previous state-of-the-art post-hoc SAE method, SpIn-ViT achieves 8.84% higher average classification accuracy, an AI-based interpretability score nearly four times as high, and a human-evaluation score more than twice as high. We further extract interpretable rule-sets using the SAE neurons to create neurosymbolic models which achieve 5.97% higher average classification accuracy while requiring a 58.8\% smaller rule-set than the neurosymbolic models created from the SOTA post-hoc SAE method. --- ### Distinguishing AI-Generated Music from Edited Audio as a Hard-Negative Robustness Task Authors: Alexandru-Stefan Morosanu, Valerian Cecan, Stefan-Daniel Achirei, Laura Erhan Published: 2026-08-14 Categories: cs.SD, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.14916v1 AI-generated music detectors are commonly evaluated against original songs, but real-world uploads are often remixed, re-encoded, pitch-shifted, or otherwise edited. These edited versions form a difficult negative class: they are not generated by AI, yet they may introduce spectral artifacts that resemble synthetic audio fingerprints. We study this problem as a hard-negative robustness setting for AI-generated music detection, focusing on AI-generated and edited variants derived from the same anchor songs. We compile a YouTube-based dataset of AI, edited, and original variants, using the original tracks only as references, and train a binary AI versus edited detector. Audio is processed as 10-second clips and passed as raw waveforms to a pretrained PaSST spectrogram transformer. To reduce leakage, all splits are performed by anchor song. On the held-out test set, the final video-level system achieves 0.811 balanced accuracy. At clip level, AI-generated clips reach an F1-score of 0.836, while edited clips reach a lower F1-score of 0.720. The results suggest that AI-generated music retains detectable fingerprint-like spectral cues beyond ordinary editing, but the lower edited-class F1-score shows that these cues can still overlap with artifacts from edited audio. Grad-CAM visualizations are used to inspect whether high-confidence predictions rely on localized time-frequency regions. --- ### The Open-Strategy Dictator Game: Cooperation Under Mutual Transparency Authors: Michael Glass Published: 2026-08-14 Categories: cs.GT, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2608.14913v1 We introduce the Open-Strategy Dictator Game (OSDG), a variant of the classic dictator game in which each player's strategy is a natural-language document visible to all participants. The dictator's decision, to SHARE or TAKE an endowment, may depend on the text of the recipient's strategy. A large language model adjudicates each interaction by interpreting the dictator's strategy in the context of the recipient's. We run round-robin tournaments among diverse strategies and analyze the resulting payoff matrix using softmax equilibrium frequencies, dominance analysis, and sensitivity to the relative value of cooperation. Conditionally cooperative strategies, those that share with cooperators and take from exploiters, consistently dominate, while unconditional strategies (always share or always take) are weakly dominated. The results suggest that in environments where agents can inspect each other's decision procedures, conditional cooperation is evolutionarily robust across a wide range of payoff parameters. --- ### Optimal Watermark Localization in Mixed-Source Large Language Model Texts Authors: Jose H. Blanchet, T. Tony Cai, Xiang Li, Hao Liu, Qi Long, Weijie J. Su Published: 2026-08-14 Categories: stat.ME, cs.CL, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2608.14906v1 Watermarking provides a principled way to authenticate text generated by large language models (LLMs). In practice, however, the final text may be mixed-source, with watermark evidence surviving at only a subset of token positions after rewriting, insertion, deletion, or paraphrasing. Although prior work has studied global detection of watermark signals, when such signals can be localized remains unclear. We formulate watermark localization as a token-level multiple-testing problem based on pivotal statistics, with a latent indicator recording whether watermark dependence survives at each position. Under an asymptotic regime indexed by exponents for signal sparsity, next-token concentration, and effective-vocabulary growth, we derive a sharp boundary for global detection and phase transitions for discovery and classification within the class of coordinatewise pivot-based localization rules. We show that discovery is strictly harder than detection and that consistent classification is impossible across the parameter regime within this class. We then develop an adaptive thresholding method that does not require knowledge of the exponents or time-varying next-token distributions, but uses a data-driven estimate of the surviving watermark fraction. The method attains the optimal discovery boundary and near-optimal discovery power relative to homogeneous pivot-based rules. Simulations support the theoretical phase transitions, while experiments on model-generated texts demonstrate practical localization performance under common edit mechanisms. --- ### How Do Agents Fail on AutoResearch: End-to-End Diagnostic Evaluation on 100 Real-World Frontier Research Tasks Authors: Yanlin Fei, Nazhou Liu, Xinmiao Yu, Shaolong Chen, Lei Li, Rahul Thapa, Madalina Ciobanu, Navan Preet Singh, Qingqing Mao, Ritankar Das Published: 2026-08-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.14905v3 AI has long assisted scientific research, but the rapid advance of LLMs and agentic scaffolds is reshaping the landscape; a single system can now carry whole-stage research from an initial hypothesis all the way to final published paper, which is a paradigm now referred to as AutoResearch. Existing evaluations reveal little about how these agents operate or where they break down. Tasks are narrowly-scoped, evaluation measures performance but not process, and failure diagnoses lack systematic coverage or artifact-level visibility. To address this gap, we introduce AutoResearchEval, featuring 100 tasks grounded in published frontier science across 7 scientific domains and the full research lifecycle, including ideation, retrieval, execution, analysis, writing, and review. Evaluating 8 harness-model combinations yields 800 autoresearch agent trajectories, with process-level annotation. We organize these insights into AutoResearch Failure Taxonomy or ARFT, a framework of 45 empirically-grounded failure patterns. To enable scalable fine-grained attribution, we leverage a human-calibrated agent-as-a-judge pipeline to inspect complete trajectories and intermediate artifacts. Failure patterns converge on a single overarching limitation, namely that current agents lack a metacognitive loop, which entails the ability to check what they produced against what they found, revise when it does not hold up, and question whether the path they took was sound. The same patterns recur across all 8 harness-model combinations, including the strongest models tested, locating the deficit at the model level rather than in any particular scaffold; whether orchestration-level interventions can close it is an open question this work does not test. We publicly release AutoResearchEval and ARFT to facilitate continued research and development in autonomous scientific discovery. --- ### Frontier AI Forecasting Has a Measurement Problem: An Audit of Progress Evidence Authors: Fabricio F Costa Published: 2026-08-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.14903v1 Quantitative forecasts of frontier artificial intelligence often connect dated targets to trends in benchmark scores, training compute, release time, or expert belief. This paper audits whether the public measurement record supports those connections before another trend is fitted. I construct a frozen, event-centric record through 12 August 2026 with 62 selected systems, 12 versioned benchmarks, seven capability or impact criteria, 144 graded events, 27 source records, and 408 typed relations. The record is an audit sample, not a census. Only seven systems jointly observe estimated training compute and a METR 50 percent task horizon. Training compute is absent for 19 of 27 closed systems, including every selected closed release from 2026, while none of the 35 open-weight systems has a METR horizon observation. Benchmark succession creates a second break: a seven-system link from METR Time Horizon 1.0 to 1.1 has a log-scale slope of 1.206 (95 percent CI 1.021 to 1.390), whereas a six-system MMLU to MMLU-Pro comparison appears shift-like under logit and probit links but not under linear or logarithmic links. The observed bridges have about 80 percent power only for slope departures near 25 percent. Provenance is concentrated: 52 of 71 substantive quantitative events, or 73.2 percent, come from one measurement programme, and 76.1 percent are laboratory releases. A review of 56 methodological and empirical sources identifies 16 complementary measurement directions spanning resources, inference budgets, reliability, agentic work, safety, human preference, field outcomes, and forecast backtesting. No direction supplies a replacement scalar. The result is not that frontier AI forecasting is impossible, but that a defensible dated forecast is a claim about a versioned measurement system with explicit joins, protocols, links, and source dependence, not merely a fitted curve or calendar date. --- ### Interpretable Cross-Lingual Alignment in Small Language Models: Probing Cultural and Pragmatic Reasoning in Japanese-English Bilingual LLMs Authors: Florian Braun Published: 2026-08-14 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.14896v1 Large language models work well on English and behave in poorly understood ways on languages typologically far from it. Japanese is a clean example, where evaluation still leans on translation quality and JGLUE-style benchmarks, which roll lexical, syntactic and pragmatic competence into a single score. The phenomena on which general-purpose models fail Japanese users are pragmatic: honorifics, in-group and out-group reference, context-sensitive politeness, zero anaphora. I introduce J-PragEval-v0, a minimal-pair benchmark isolating four such phenomena from surface fluency, and combine it with linear probes and teacher-forced log-probability evaluation to ask where inside TinySwallow-1.5B (28 layers, hidden size 1536) the corresponding contrasts live. The four features split three ways. Honorific register sits cleanly in the residual stream: 0.96 balanced accuracy at layer 15, and the model flips its preferred continuation with the scenario on 93 percent of items. Implicit subject and in-group reference are not linearly decodable at the final prompt token (0.48 and 0.38), yet flip rates are 0.77 and 0.79, so the contrast is worked out during generation rather than stored at the prompt. Indirect refusal is the negative case: 0.95 probe accuracy collapsing to a 0.43 flip rate under length-normalised teacher forcing, because the current minimal pairs conflate politeness with continuation length. I also specify Pragmatic Representation Steering, a parameter-free inference-time method that edits residual-stream activations along the class-mean-difference directions probing identifies. Feasibility is argued indirectly rather than demonstrated: the contrastive activation addition baseline, the same geometry the method would inject, recovers probe accuracy within one to two points of logistic regression wherever a linear signal exists. Scaling to Llama-3.1-Swallow-8B is the next step. --- ### Can Neural Networks Learn by Experimenting on Themselves? Self-Interventional Learning from Functional Consequences to Predictive Self-Knowledge Authors: Michał Tomaszewski Published: 2026-08-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.14894v1 Machine-learning systems usually model external data, while their internal functional organization is analyzed by external observers. This work introduces Self-Interventional Learning (SIL), in which a neural system perturbs its own functional structure, observes consequences, learns a predictive self-model, generalizes to unexecuted interventions, and uses predictions to guide later structural action. In a construction-known synthetic system, SIL recovered critical structure, redundancy, and replaceability, while synergy was not reliably recovered. Across 30 fresh confirmatory seeds, increasing the pairwise intervention budget from 4 to 56 reduced held-out prediction error from 0.0335 to 0.0148 and increased Spearman correlation from 0.629 to 0.883. In a matched ablation, preserving the correct intervention--consequence mapping reduced prospective prediction error by 81.3%, while using the same learned self-model for action reduced normalized regret by 31.7% relative to ignoring it. However, model-guided action did not significantly outperform a direct empirical-memory policy, and powered CIFAR-10/ResNet validation showed no robustness advantage over equal-budget direct repair search. These results support SIL as an intervention-driven framework for learning predictive knowledge about a network's own functional organization, while showing that the self-model remains incomplete and is not universally superior to simpler direct strategies. --- ### Where Does Retrieval Fail? Evaluating RAG Architectures for Agricultural Advisory Authors: Khan Raiyan Ibne Reza, Sanjana Aktar Maria, Sumaiya Tabassum Nimi Published: 2026-08-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.14886v1 Retrieval quality in RAG systems is commonly reported as a single aggregate score, which can hide large differences across query types and language conditions. We study this problem in Bengali agricultural advisory, where farmer queries are often colloquial while official advisory documents use formal scientific terminology. We construct a test collection of 1,000 queries and 2,882 knowledge nodes extracted from 284 official Bangladeshi agricultural publications, and use it to evaluate five retrieval architectures and six embedding models under three controlled language conditions. The results show that no single retrieval method is consistently best. For native Bengali queries, BM25 is the strongest single retriever (R@10 = 0.506) while Hybrid RRF reaches the highest overall R@10 of 0.539. However, dense retrieval performance varies sharply by query type: R@10 is 0.093 on colloquial farmer queries and 0.970 on formal safety queries. Across language conditions, BM25 R@10 drops from 0.506 on Bengali queries to 0.004 when English queries are matched against the Bengali corpus, while dense retrieval falls only from 0.464 to 0.425. We also find that embedding task configuration and passage length can each change reported R@10 by a factor of seven, independent of architecture. These results show why low-resource RAG evaluation should report performance by language condition and query type rather than relying on aggregate scores alone. The dataset and evaluation scripts are available at https://huggingface.co/datasets/RaiyanKhaan/AgriTrust-RAG. --- ### Personalized Auto-Research: Towards a True AI Co-Scientist Authors: Bo Ni, Franck Dernoncourt, Hongjie Chen, Yu Wang, Nesreen K. Ahmed, Zhengzhong Tu, Tyler Derr, Ryan A. Rossi Published: 2026-08-14 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2608.14881v1 AI co-scientists that generate hypotheses, retrieve related work, design experiments, execute code, and draft full papers are beginning to change how research is carried out. Despite this rapid progress, state-of-the-art systems remain researcher-agnostic: given a research goal, they optimize novelty, validity, or reviewer score while ignoring the individual scientist who will use the output. This overlooks a fundamental fact about research, namely, that what counts as novel, valuable, or feasible depends on the researcher, including their prior work, methodological repertoire, and the collaborators and communities in which they are embedded. In this work, we introduce the problem of personalized auto-research, which conditions every stage of the research process on a representation of the individual researcher. We argue that personalization is not a convenience layer, but rather the fundamental property that allows an AI system to serve as a genuine co-scientist rather than a generic instrument. To address this problem, we propose a general and flexible framework that threads a graph-grounded researcher context through retrieval, hypothesis search, experimentation, writing, and review. The framework consists of three fundamental components: (i) graph-grounded researcher representations, (ii) personalization across the full research pipeline, and (iii) evaluation grounded in the individual. Notably, we highlight a one-size-fits-all failure mode where distinct researchers issuing the same goal receive essentially the same research, erasing the tacit knowledge through which novel ideas arise. Finally, we discuss fundamental open problems and challenges. --- ### Workspace Topology as an Attack Vector in Agentic Coding Assistants Authors: Alexandre G. R. Day, Pradeep Yadlapalli, Sriram Venkatapathy, Thomas Paniagua, Nick Raines, Sahil Wadhwa, Himanshu Kumar, Andy Luo, Sudeep Panyam, Rikhiya Ghosh, Pranab Mohanty, Giri Iyengar Published: 2026-08-14 Categories: cs.CR, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.14876v1 Agentic coding assistants are finding widespread use, not just in new code development but in quickly ingesting and leveraging third-party code. This opens up a risk of malicious code being ingested as these coding tools operate with broad filesystem access inside developer workspaces. In this paper, we extensively study the impact of different dimensions of a novel attack surface we term workspace topology -- defined via directory depth, codebase modularity, in-file injection position and context framing -- on the attack success rate of adversarial prompt injection attempts. We perform an empirical study of indirect prompt injection (IPI) across a diverse set of open-source repositories spanning 10 languages and 6 engineering domains, evaluating three IPI entry points against open-weight models operating open source code harnesses. We find that workspace topology measurably affects IPI success. Specifically, changes in codebase modularity can significantly alter the Attack Success Rate (ASR), with highly modular environments demonstrating significantly lower attack success rates. Furthermore, context framing and introduction of security-cues in the workspace can also alter the ASR. Our findings offer practical value for the evaluation and security testing of coding agents across diverse settings, while underscoring the importance of an uncontaminated testing environment to obtain reliable results and conclusions. --- ### When do machine-learned exchange-correlation improvements inherit into density-functional tight binding? Authors: Can Polat, Mustafa Kurban, Erchin Serpedin, Hasan Kurban Published: 2026-08-14 Categories: cond-mat.mtrl-sci, cs.LG, physics.chem-ph, quant-ph arXiv: https://arxiv.org/abs/2608.14875v1 Machine-learned exchange-correlation functionals correct band gaps at near-semilocal cost, while density-functional tight binding reaches the $10^3$-$10^6$-atom regime; combining them assumes that a better parent yields a better parameterization, but we show it does not. Current-generation functionals are orbital-dependent generalized Kohn-Sham operators, whereas the parameterization channel is built on a multiplicative potential, preventing exact representation. Using the transfer ratio, the surviving fraction of a parent-level change, we find anti-transfer: coherently negative ratios across four covalent semiconductors move the gap in the wrong direction, consistent with a molecular proxy and an r$^2$SCAN control. The minimal-basis overgap is dominated by the on-site convention rather than basis incompleteness; correcting the on-site block removes most of it, while one $d$-polarization shell closes a further $16$-$40%$, depending on the placement of the empty $d$ level, which no free-atom eigenvalue uniquely fixes. Occupied-manifold enhancements, ionic and closed-shell repulsive potentials, and rocksalt-oxide gaps inherit, whereas elemental and III-V covalent networks inherit neither gaps nor repulsive potentials and oxide networks inherit only the latter. We screen 23 elements and release the parameter sets, showing that the transfer ratio provides a cheap pre-test before any parameterization campaign. --- ### JarvisBench: Always-on Intelligence Between Humans and Agents Authors: Chen Chen, Zhehuai Chen Published: 2026-08-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.14870v1 Long-horizon agents can execute continuously, but human attention remains intermittent and scarce. This creates a bidirectional coordination problem: users may need immediate access to an agent while work continues in the background, whereas agents may encounter consequential decisions that require user judgment after the user has stopped monitoring execution. We posit an always-on attention-coordination layer---\textit{Jarvis}\footnote{Named after the fictional AI assistant in \textit{Iron Man}.}---that mediates this interface and allocates human attention across one or more working agents. We introduce \textit{JarvisBench} to evaluate both directions of this coordination: whether an intermediary can accurately and promptly answer user-initiated questions about ongoing work, and whether it can recognize when an agent requires user judgment, solicit that judgment at the right moment, and route it back to improve task outcomes. JarvisBench contains 45 agentic task instances: 20 single-agent tasks and 25 workstreams organized into 10 multi-agent projects. The tasks span 19 domains and were selected and adapted from more than 2,000 public candidates. Crucially, the need for user attention arises naturally during execution rather than from an obvious omission in the initial prompt. JarvisBench is designed to integrate with arbitrary agent runtimes without modifying their underlying execution loops. Our reference implementation further provides a full-duplex speech interface, allowing users to reach Jarvis naturally while timely attention coordination supports agents working in the background. By separating agent execution from attention coordination, JarvisBench provides a stable evaluation target as agent capabilities continue to improve. --- ### Beam-Wise Statistical Background Subtraction for Static Roadside LiDAR: A Cross-Sensor Benchmark Study Authors: Alexander Baumann, Marcel Vosshans, Thao Dang Published: 2026-08-14 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2608.14868v1 Background subtraction is a key preprocessing step for infrastructure-based LiDAR perception, enabling efficient isolation of dynamic traffic participants without semantic annotations. However, systematic cross-sensor evaluations and reproducible studies for static roadside LiDAR are missing. This paper presents a comparative benchmark of beam-wise statistical background subtraction for statically mounted LiDAR sensors. We formulate background estimation as a per-beam temporal modeling problem and investigate complementary statistical strategies that capture dominant as well as multi-modal background structures, combined with spatial filtering in the angular and 3D domain. To enable reproducible evaluation, we introduce HighwayScene, a new multi-LiDAR dataset recorded in a static roadside setup, and extend the public CoopScenes dataset with static/dynamic point-wise annotations. Across multiple scenes and heterogeneous sensing technologies, we demonstrate that beam-wise statistical modeling provides a robust and transferable solution. Combining lightweight per-beam models with spatial consistency filtering substantially improves precision while maintaining high recall and real-time capability. All datasets, annotations, and implementations are publicly released. --- ### ARISE: An adaptive residual-informed stability ensemble for feature selection in small-sample biomedical omics Authors: Zardad Khan, Amjad Ali, Naz Gul, Sheema Gul, Saeed Aldahmani Published: 2026-08-14 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2608.14866v1 Objective: Small-sample molecular classification requires feature selectors that identify predictive, stable, and nonredundant subsets for binary and multiclass outcomes. We propose ARISE (Adaptive Residual-Informed Stability Ensemble), which integrates complementary relevance signals, class-balanced stability assessment, residual-informed redundancy control, and multiclass pairwise coverage. Methods: ARISE combines seven percentile-normalized relevance components through 15 predefined profiles, adaptively weighted by nested inner cross-validation. It was evaluated on five molecular datasets, eight feature-set sizes, three fixed classifiers (k-nearest neighbours, support vector machine, and random forest), and six filter comparators. Generalization was estimated by five-fold outer cross-validation repeated 50 times using balanced accuracy, macro-F1, and Cohen's kappa. Results: Across 210,000 held-out assessments, ARISE ranked first in all 15 dataset-metric combinations. Equal-dataset means were 0.793 for balanced accuracy, 0.776 for macro-F1, and 0.725 for kappa, exceeding the strongest aggregate comparator by 0.022, 0.023, and 0.028, respectively. Performance remained strong across compact feature sets, although the optimal budget differed by dataset. Conclusion: ARISE provides a transparent, adaptive framework that jointly addresses relevance, stability, redundancy, and multiclass discrimination. Its consistent results across datasets, classifiers, metrics, and feature-set sizes support further evaluation for small-sample molecular classification. --- ### Evaluating Agentic Code Repair Capabilities in Distributed Systems Authors: Yibo Yan, Huijuan Wang, Junzhou He, Yizhuo Liang, Shaoyu Wang, Huanchen Sun, Seo Jin Park Published: 2026-08-14 Categories: cs.SE, cs.AI, cs.DC arXiv: https://arxiv.org/abs/2608.14863v1 LLM-based coding agents have advanced rapidly on single-process SWE tasks, with frontier models now clustering in the high-70s on SWE-bench Verified. Distributed-system debugging, however, remains an under-explored regime: bugs span processes, nodes, and protocol interactions, with root causes rarely recoverable from source alone and brute-force exploration intractable across non-deterministic interleavings. This leaves two gaps in LLM and agent evaluation: no code-repair benchmark targets distributed-system bugs, and no controlled study isolates how much externally provided debugging context changes agent success on them. We introduce DDBench, a code-repair benchmark of 60 historical bugs mined from 13 open-source distributed systems, partitioned into three difficulty tiers. DDBench evaluates every case under two matched conditions: a symptom-only condition where the agent receives only the bug symptom and repository, and a context-augmented condition where it additionally receives a bounded debugging context (logs, traces, runtime state, and targeted code-investigation notes), isolating the effect of debugging context from model capability. The evaluation of ten LLMs on DDBench reveals several findings. First, distributed debugging exercises a reasoning dimension that single-process benchmarks do not surface: models' pass rates span 61 pp, and pairwise bootstrap separates 9 of 15 top-tier model pairs at p < 0.05 on DDBench's hardest case-set. Second, bounded debugging context lifts aggregate pass rate by +18.1 pp, and the lift is asymmetric: weaker models gain pass rate, while stronger models gain efficiency. Third, debugging context requires careful curation, as even faithful debugging context can sometimes mislead LLMs. --- ### STAR-FL: Secure Federated Learning with Spatial-Temporal Analysis and Robust Aggregation Authors: Nawrin Tabassum, Yanzhao Wu Published: 2026-08-14 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2608.14861v1 Data poisoning attacks pose serious security threats to Federated Learning (FL) systems in Computer Vision. Despite growing research attention, two key challenges remain for existing defense techniques: (1) accurately distinguishing between benign and malicious model updates and (2) effectively mitigating the influence of poisoned model updates during model aggregation. To address these challenges, we propose a novel defense framework against targeted poisoning attacks with Spatial-Temporal Analysis and Robust aggregation for FL (STAR-FL). First, we employ spatial-temporal clustering to identify and remove potentially malicious updates from the FL training process. Second, we adjust the learning rate during aggregation to mitigate the impact of any malicious updates that evade detection. Third, we conduct extensive experiments across multiple benchmark datasets to evaluate the spatial-temporal analysis and robust aggregation in STAR-FL. Experimental results demonstrate their synergistic effect in enabling STAR-FL to effectively protect FL and consistently outperform state-of-the-art defenses against targeted poisoning attacks, significantly reducing Attack Success Rates (ASRs). The source code is available at https://github.com/mlsysx/STAR-FL. --- ### BRAID: Learning Equilibrium Maps in Interdependent Security Games via Weight-Tied Iterative Graph Neural Networks Authors: Elnaz Nowrouzi, Zhiqun Zuo, Xueru Zhang, Mohammad Mahdi Khalili Published: 2026-08-14 Categories: cs.GT, cs.LG arXiv: https://arxiv.org/abs/2608.14856v1 Computing Nash equilibria in interdependent security (IDS) games on networks is computationally expensive: best-response dynamics may need hundreds of iterations per instance, and downstream tasks such as auditing, stress-testing, and incentive design often require repeatedly re-solving the game under parameter perturbations. We propose BRAID, a Best-Response Amortized Iterative Dynamics model that uses a weight-tied iterative graph neural network to learn a direct map from game parameters to Nash equilibrium effort profiles, replacing iterative best response computation with a single forward pass that is up to 43X faster per instance. BRAID is derived from the best-response fixed-point structure of IDS games: its SUM aggregation reflects additive neighbor coupling, and a weight-tied gated recurrent unit (GRU) mirrors a damped best-response update. The same architecture applies across IDS specifications that vary investment-cost curvature and neighborhood aggregation, including log-linear, quadratic-cost, and log constant-elasticity-of-substitution (CES) utilities. Beyond equilibrium prediction, BRAID also recovers how equilibrium efforts change under perturbations to game parameters, including costs and network edge weights. We make this sensitivity recovery an explicit evaluation target and introduce two training strategies, interior-equilibrium training and input-noise regularization, that improve the local behavior of the learned equilibrium map without using sensitivity labels. Experiments show that BRAID effectively predicts Nash equilibria and recovers equilibrium sensitivities across utility specifications and network sizes. --- ### What to Forget in Unlearning? Forget Set Curation for Language Models Authors: Animesh Jha, Arpandeep Khatua, Youssef Allouah, Sanmi Koyejo Published: 2026-08-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.14855v1 Machine unlearning aims to remove targeted data or behaviors from a trained model without retraining from scratch. Yet most evaluations assume that the examples to forget are already known. In realistic language-model deployments, a requester may ask a model to stop reproducing a song or book without knowing which spans, documents, quotations, or near-duplicates in a trillion-token corpus support that behavior. We study this missing upstream problem, forget set curation: mapping a suppression request to the data passed to an unlearning algorithm. We introduce CleanSlate, a benchmark for verbatim output suppression over songs and books, with model-specific extraction profiles, content-grounded QA, and capability-retention evaluations. CleanSlate exposes two failure modes. Natural lexical and exact-substring curators often yield forget sets that lead to weak suppression. An evaluation-aware curator suppresses requested continuations almost completely, but causes collateral regression on non-requested content and model-dependent capability loss. These results show that practical unlearning is not only an optimization problem once a forget set is given: the data chosen for forgetting determines both what can be unlearnt and what else is damaged. --- ### Zero-MELO: Test-Time Evidence Calibration with Multimodal LLMs for Zero-Shot Micro-Gesture Recognition Authors: Chengyan Wang, Hanliang Xie, Yueyi Yang, Haoyu Chen Published: 2026-08-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.14854v1 While Multimodal Large Language Models (MLLMs) excel in general video understanding, their capability in fine-grained and motion-centric tasks remains limited. This limitation is particularly critical in micro-gesture recognition (MGR), where micro-gestures (MGs) - subtle, short-duration, and spatially localized human movements - serve as key discriminative signals for implicit affective analysis, yet are easily neglected following common prompting practices. Although MGR has been intensively studied by many discriminative approaches, the use of MLLMs for MGR is underexplored, with notably poor performance. We hypothesize that the motion-sensitive representation ability of MLLMs is constrained by their inherent single-pass forward inference, which can be substantially enhanced through carefully designed test-time guidance. Motivated by this, building on our prior findings regarding temporal insensitivity in Video LLMs, we diagnose zero-shot MGR errors in the Negative Log-Likelihood (NLL) space. We observe that MLLMs suffer from two bottlenecks: 1) insufficient localized evidence and 2) severe score biases driven by language and motion-agnostic appearances. Thus, we propose a novel test-time evidence calibration framework that improves both reasoning details and prediction reliability. Specifically, we introduce a tree search mechanism to progressively acquire localized, fine-grained visual evidence, coupled with a test-time calibration module to mitigate score biases. The multi-cue fusion module then integrates evidence from multiple cues without relying on a single cue for final prediction. Our framework achieves mean-class accuracies of 26.84\% on iMiGUE and 22.10\% on MA-52, significantly outperforming the Qwen2.5-VL baseline, which produces 16.15\% and 10.20\%, respectively. The code will be available at https://zero-melo.github.io/Zero-MELO. --- ### Discovering High-Quality Chess Puzzles with Offline Reinforcement Learning Authors: Allen Nie, Anirudhan Badrinath, Nicholas Tomlin, Timothy Dai, Carissa Yip, Rose E Wang, Emma Brunskill, Chris Piech Published: 2026-08-14 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.14851v1 Learning and skill mastery require extensive and deliberate practice. In many learning settings, producing high-quality pedagogical materials can require a high level of domain expertise and be very time-consuming. Pedagogical materials often need to train students to engage in different thinking patterns. In some domains, such as chess, puzzles are used to help students practice their skills in calculating the next moves and recognizing known patterns on a board. Giving students a practice set of puzzles to help them learn different modes of thinking is challenging because the teacher needs to carefully balance between different motifs and how many look-ahead steps a student needs to perform. Popular online platforms like Chess.com and Lichess offer players millions of puzzles. Unlike chess tactics puzzles procured by human experts, where chess beginners can learn valuable insights, these puzzles are automatically generated and often regarded as having low pedagogical value. These platforms also rely on a heuristic to recommend puzzles to users for practice. Using the user history data over an entire year, a total of 1.5 billion puzzle-solving histories, we learn the pedagogical value of a puzzle and how to automatically choose a set of puzzles to better support chess learners using insights from offline reinforcement learning. We show that using offline policy evaluation, our trained policy has significant impact on beginners with puzzle-solving Elo range of 100--1000, particularly for the group of beginners whose learning growth was stagnant. We also performed a qualitative analysis of the puzzles discovered by our model by collecting annotation ratings from expert chess players. The success of our pipeline shows promise for a future where we can understand the pedagogical values of practice items given general user interaction data. --- ### M-LINKX: Multiview Graph Learning for Brain Cognitive Disease Detection Authors: An Phan, Yufei Jin, Xingquan Zhu Published: 2026-08-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.14847v1 Electroencephalogram (EEG) is a non-invasive and relatively low-cost procedure that measures brain electricity for the detection of cognitive diseases. EEG-based classification of dementia-related conditions, including Alzheimer's disease (AD), mild cognitive impairment (MCI), and frontotemporal dementia (FTD), remains challenging because EEG signals are noisy, non-stationary, and vary across subjects. Segment-based learning provides a practical way to model long EEG recordings by converting them into fixed-length inputs. For each segment, discriminative information may be explored by using signals within each channel (i.e. electrode), as well as interactions between EEG channels. In this paper, we propose M-LINKX, a multi-view graph learning framework for EEG-based dementia classification. For each segment, we extract channel-level node features and construct multiple functional-connectivity (FC) graph views, where each view is defined by a specific combination of connectivity metric, frequency band, and topology filter, respectively. Instead of relying on message passing over the constructed graphs, M-LINKX follows a simple design in modeling node features and adjacency-based connectivity representations. The graph-view representations are fused using global trainable view weights, and subject-level prediction is obtained by averaging segment-level probabilities. Experiments on two three-class EEG datasets with different diagnostic groups, CAUEEG (HC/MCI/Dementia) and AHEAP (HC/AD/FTD), show that M-LINKX achieves the best subject-level performance under the main experimental settings. Our study suggests that multi-view functional connectivity can improve EEG-based dementia classification when integrated with an appropriate graph-learning architecture. Code and data are available at https://github.com/anphantt/MLINKX. --- ### Writing Style Similarity Reflects Academic Genealogy Authors: Cameron Manzo Published: 2026-08-14 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.14843v2 As authorship attribution systems are increasingly deployed to detect ghostwritten and AI-generated papers, their errors can support accusations against legitimate authors. These systems conflate stylistic similarity with individual identity. Researchers, however, study under advisors, and inherit their stylistic quirks. We build a corpus of arXiv authors with $\geq 2$ solo papers from the Mathematics Genealogy Project graph, giving $5{,}803$ total authors and $2{,}501$ ground-truth advisor-student pairings. Using embeddings from a fine-tuned model, advisors sit $39.9\%$ closer in cosine distance to their students than a random same-field author does. Using two open models, we reproduce the effect at $12.6\%$ and $14.5\%$. Academic siblings, two students of one advisor who may never have met, sit $30.4\%$ closer across $8{,}360$ pairs, even when they studied at different institutions. Pairs who share only institution and field show negligible similarity. Given a closed-set attribution task over the same corpus, the system's errors occur on the true author's advisor, student, academic sibling, or lab mate $11$ times more often than chance. --- ### What the Reranker Sees: Multi-Aspect Page Annotation for Long-Document Multimodal Question Answering Authors: Guanchen Wu, Jiayuan Ding, Subhabrata Mukherjee, Carl Yang Published: 2026-08-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.14841v1 Long-document visual question answering (VQA) over documents of tens to hundreds of pages mixing text, tables, charts, and figures typically follows retrieve-then-read pipelines. In our setting, the bottleneck shifts from retrieval recall to reranker-side evidence selection: on MMLongBench-Doc, BGE-M3 reaches Recall@20 = 0.86 but only F1@5 = 0.254, and even the visual retriever ColPali reaches only F1@5 = 0.332; a text-only rerank LLM seeing only raw snippets misses table, chart, and layout evidence even when the upstream retriever encoded images. We propose Trident, with two complementary components: Trident-R, a retriever-agnostic LLM reranker that converts each candidate into an LLM-readable semantic record, including a visual caption, section path, entity tags, multi-axis concept hits, and a text snippet, then performs a single adaptive-K rerank call; and Trident-S, a generation-side module that prompts the VLM under topical, entity, and structural lenses before synthesis. On two long-document datasets, the annotation+rerank protocol substantially improves retrieval F1 across five heterogeneous pools, with every reranked pool exceeding the strongest adaptive-K baseline PageIndex. An LLM rerank without the annotation barely changes first-hit ranking, indicating the lift comes from the structured annotation. Trident-S targets open-ended synthesis questions by design, adding up to 6.6 points in generation accuracy on these questions. The best Trident configuration is the strongest downstream QA pipeline in our evaluation, with rankings consistent across two LLM judges (kappa = 0.913). ---