# Research Papers — 2026-05 · bundle page 19 of 326

> Full-text markdown of papers #901–950 of 16266 published in 2026-05, newest first. AI-readable mirrors of the full 16266-paper 2026-05 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-05/page/19
[← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/18.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-05/page/20.md)

### TunerDiT: Training-free Progressive Steering of Diffusion Transformer for Multi-Event Video Generation

Authors: Ruotong Liao, Guowen Huang, Qing Cheng, Guangyao Zhai, Lei Zhang, Xun Xiao, Thomas Seidl, Daniel Cremers, Volker Tresp
Published: 2026-05-29
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2605.31590v1

Text-to-video (T2V) generation faces challenging questions when generating videos with long horizons containing multiple events. Inspired by the intrinsics of the diffusion process, we probe video diffusion transformers (DiTs) and uncover intrinsic turning points in the DiT denoising trajectory where conditioning text affects generation from global layout to fine-grained details. Building on this finding, we present TunerDiT, a simple yet effective progressive steering method that requires no additional training for multi-event generation. TunerDiT comprises two steering handles: (1) Event-Partitioned Masking that enforces event boundaries while allowing cross-event transition bands; (2) Cross-Event Prompt Fusion that injects neighboring event semantics for late-stage refinement. We contribute a self-curated prompt suite for benchmarking multi-event generation, i.e., Meve. TunerDiT achieves state-of-the-art performance across 8 metrics and offers a tunable trade-off between video consistency and event separation, compared with other training-free methods. The improvement in text alignment increases with the event count, indicating a scaling possibility with increasing event count.

---

### Recognizing Co-Speech Gestures in-the-Wild

Authors: Sindhu B Hegde, K R Prajwal, Andrew Zisserman
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31589v2

While humans naturally gesture during speech, only a sparse subset of these co-speech gestures are visually depictive and semantically linked to specific spoken words. In this paper, we introduce a large-scale dataset -- Gesture Recognition in the Wild (GRW), comprising co-speech gestures corresponding to a diverse vocabulary of 155 words. GRW contains 140k manually annotated video clips where the word is spoken, with 17k instances of semantic co-speech gestures including their frame-level temporal boundaries. The video clips are collected 'in the wild' from public-facing discourse, including lectures, talk shows, and interviews, covering a diverse range of speakers and visual conditions. We also introduce video models to: (a) classify gestures as semantic or not; (b) recognize the word corresponding to a co-speech gesture; and (c) temporally localize the gesture. These models are trained and evaluated on the GRW dataset and compared against a range of strong baselines, establishing benchmark results for all three tasks. The dataset, annotations, and trained models are publicly available on the project website: https://www.robots.ox.ac.uk/~vgg/research/grw.

---

### Recognizing Co-Speech Gestures in-the-Wild

Authors: Sindhu B Hegde, K R Prajwal, Andrew Zisserman
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31589v1

While humans naturally gesture during speech, only a sparse subset of these movements are visually depictive and semantically linked to specific spoken words. Current multimodal models struggle to capture these semantic co-speech gestures, heavily bottlenecked by a lack of precisely annotated training data. To address this, we introduce the Gesture Recognition in the Wild (GRW) dataset, the first large-scale benchmark designed to map unconstrained human gestures to specific words with frame-accurate temporal boundaries. Comprising 156,688 manually annotated video clips, GRW spans a highly diverse 150-word taxonomy of physical actions, spatial descriptors, and abstract concepts. We leverage GRW to train video models to (a) classify gestures as semantic or not, (b) recognize the word corresponding to a co-speech gesture, and (c) temporally localize the gesture. We also use GRW to establish benchmarks for these three tasks.

---

### Language Models Learn Constructional Semantics, Not To Mention Syntax: Investigating LM Understanding of Paired-Focus Constructions

Authors: Wesley Scivetti, Ethan Wilcox, Nathan Schneider, Kanishka Misra, Leonie Weissweiler
Published: 2026-05-29
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2605.31586v1

Grasping the semantics of rare constructions (form-meaning pairings) has been shown to be a challenging problem that has currently only been solved by the largest LLMs. It remains an open question if open-source models have robust constructional understanding, and if so, what learning dynamics underlie the acquisition of this knowledge. Focusing on a set of rare Paired-Focus constructions in English (e.g. "let alone", "much less"), we construct a novel dataset to test their meanings using both scalar adjectival semantics and general world knowledge. Testing a wide range of models differing in parameter count, architecture, and pretraining dataset size, we find that several modestly sized models are sensitive to both the forms and the meanings of Paired-Focus constructions, though models trained on human-scale data fail at all meaning evaluations. Turning to training dynamics for a set of open-checkpoint models, we find that Paired-Focus understanding emerges later in training than Paired-Focus syntactic knowledge, and that learning of Paired-Focus semantics is correlated with gains in some domains of world knowledge. Overall, our empirical results support the conclusion that modestly sized open-source models can grasp the rare Paired-Focus constructions, and demonstrate a connection between knowledge of Paired-Focus constructions and other meaning domains.

---

### LongTraceRL: Learning Long-Context Reasoning from Search Agent Trajectories with Rubric Rewards

Authors: Nianyi Lin, Jiajie Zhang, Lei Hou, Juanzi Li
Published: 2026-05-29
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2605.31584v1

Long-context reasoning remains a central challenge for large language models, which often fail to locate and integrate key information in extensive distracting content. Reinforcement learning with verifiable rewards (RLVR) has shown promise for this task, yet existing methods are limited by low-confusability distractors and sparse, outcome-only reward signals that cannot supervise intermediate reasoning steps. To address these issues, we introduce \textsc{LongTraceRL}. For data construction, we generate multi-hop questions via knowledge graph random walks and leverage search agent trajectories to build \emph{tiered distractors}: documents the agent read but did not cite (high confusability) and documents that appeared in search results but were never opened (low confusability), producing training contexts that are far more challenging than those built by random sampling or one-shot search. For reward design, we propose a \emph{rubric reward} that uses the gold entities along each reasoning chain as fine-grained, entity-level process supervision. This rubric reward is applied only to responses with correct final answers (positive-only strategy), distinguishing the reasoning quality among correct responses and preventing reward hacking. Experiments on three reasoning LLMs (4B--30B) across five long-context benchmarks demonstrate that \textsc{LongTraceRL} consistently outperforms strong baselines and encourages comprehensive, evidence-grounded reasoning. Codes, datasets and models are available at \href{https://github.com/THU-KEG/LongTraceRL}{https://github.com/THU-KEG/LongTraceRL}.

---

### Choosing the Lens: Strategic Perspective Activation in Context-Dependent Argumentation

Authors: Albert Sadowski, Jarosław A. Chudziak
Published: 2026-05-29
Categories: cs.AI
arXiv: https://arxiv.org/abs/2605.31581v2

The same arguments often need to be evaluated under different external regimes. An agent with influence over the regime has a strategic lever that standard formalisms do not directly capture. We introduce context-dependent argumentation frameworks (CDAFs), an extension of Dung's theory in which a defeat function determines, per context, which attacks succeed. Blocked attacks are inverted rather than deleted, so extensions stay conflict-free with respect to the attack relation. A perspective-labeled specialisation derives the defeat function from a relevance set $ρ$ and a priority $π$. The relevance set is the agent's action space. In a small worked example, the agent's target argument is rejected under every full-relevance priority, yet accepted under a partial activation whose outcome no VAF audience can mirror. We define the corresponding decision problem, ACTIVATION-MANIPULATION, and record baseline complexity bounds. For grounded semantics with mandatory perspectives the problem is NP-complete, and the hardness comes from the activation choice itself.

---

### Giving Sensors a Voice: Multimodal JEPA for Semantic Time-Series Embeddings

Authors: Utsav Dutta, Gerardo Pastrana, Sina Khoshfetrat Pakazad, Henrik Ohlsson
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.31580v1

Transformer-based architectures have advanced sequence modeling in language and vision, yet general-purpose representation learning for heterogeneous multivariate time series remains underexplored. We introduce CHARM (Channel-Aware Representation Model), which incorporates channel-level textual descriptions into a Transformer encoder equivariant to channel order. CHARM is trained with a Joint Embedding Predictive Architecture (JEPA) and a novel loss promoting informative, temporally stable embeddings; latent-space prediction encourages robustness to sensor noise while description-aware gating provides interpretability through learned inter-channel relationships. Across anomaly detection, classification, and short- and long-term forecasting, the learned embeddings achieve strong performance using only a linear probe. Performance is driven primarily by the JEPA objective and conditioning architecture, with text descriptions serving as channel identifiers for cross-dataset generalization.

---

### SurGe: Improved Surface Geometry in Point Maps

Authors: Karim Knaebel, Gonzalo Martin Garcia, Christian Schmidt, Ilya Fradlin, Lucas Nunes, Daan de Geus, Bastian Leibe
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31577v1

Recent feedforward 3D reconstruction methods predict point maps and estimate global 3D geometry remarkably well. However, their predictions still exhibit inaccurate local surface geometry, which is clearly visible qualitatively but only weakly reflected in common metrics. To make these errors more explicit in evaluation, we introduce a point map normal metric that evaluates the local surface orientation induced by neighboring 3D predictions. To reduce these errors, we propose two complementary components: a point gradient matching loss that supervises depth-normalized 3D finite differences, and a Neighborhood Attention Decoder (NAD) that progressively upsamples features and uses Neighborhood Attention for local feature mixing. Across eight zero-shot monocular geometry benchmarks, our model, SurGe, achieves the best average rank for global point map AbsRel and consistently improves local point map and point map normal evaluations.

---

### Joint Multi-Camera LiDAR Extrinsic Calibration via Learned Pairwise Initialization and Geometric Refinement

Authors: Aziz Al-Najjar, Marzieh Amini, James R. Green, Felix Kwamena
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31576v1

Most learning-based camera-LiDAR calibration methods treat each camera-LiDAR pair independently, ignoring the rigid geometric coupling in multi-camera platforms. As a result, per-camera estimates may be individually accurate yet inconsistent at the system level. We present a two-stage framework for joint multi-camera LiDAR extrinsic calibration that combines learned pairwise matching with geometric refinement. First, CMRNext is applied independently to each camera to produce initial extrinsic estimates and dense 2D-3D correspondences. These predictions are then jointly refined through a multi-frame bundle adjustment with reprojection, per-camera prior, and relative-pose prior terms. This approach converts pairwise predictions into a globally consistent multi-camera calibration. Experiments on KITTI (in-domain for CMRNext) and Walkley (out-of-domain) datasets show improved per-camera accuracy and inter-camera consistency. On KITTI, the method achieves 0.89 cm translation error and 0.038 rotation error. On Walkley, it reduces translation error from 108.6 cm to 3.1 cm, highlighting the benefit of explicit multi-camera coupling when single-camera predictions are less reliable.

---

### SPECTRA: Synthetic IR Test Collections with Relevance Oracles and Controlled Distractor Diagnostics

Authors: Eric Liang
Published: 2026-05-29
Categories: cs.IR, cs.AI
arXiv: https://arxiv.org/abs/2605.31575v1

Scalable information retrieval testing needs corpora that are large enough to stress index construction, ranking latency, query routing, and evaluation tooling, yet human-judged test collections remain expensive and may be unavailable when documents are private or still under design. This paper introduces SPECTRA, a reproducible framework for generating synthetic text corpora and retrieval test collections through a separation of latent topical structure, surface text realization, metadata controls, query intent generation, and deterministic relevance oracles. The framework is intended as a diagnostic complement to Cranfield-style and TREC-style evaluation, not as a replacement for human assessment. A single-process Python prototype generated corpora up to 60,000 documents and 9.61 million tokens while preserving controllable long-tail vocabulary growth and producing graded relevance labels for 96 queries. In the local simulation study, generation remained close to linear at roughly 12K to 14K documents per second, estimated Zipf slopes stayed near 0.86 in absolute value, and increasing cross-topic distractor text reduced BM25 nDCG@10 from 1.00 at 2% distractors to 0.43 at 36% distractors. These results show that lightweight synthetic corpora can expose retrieval-system scaling and failure modes before costly collection construction begins.

---

### nuReasoning: A Reasoning-Centric Dataset and Benchmark for Long-Tail Autonomous Driving

Authors: Zhiyu Huang, Johnson Liu, Rui Song, Zewei Zhou, Ruining Yang, Yun Zhang, Tianhui Cai, Hanyin Zhang, Mingxuan Gao, Valeria Xu, Jiali Chen, Yishan Shen et al.
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31572v1

Reasoning is essential for autonomous driving (AD) in long-tail scenarios, where vehicles must apply commonsense knowledge, understand spatial relations, infer agent interactions, and make safe decisions. However, existing AD datasets and benchmarks mainly target perception, prediction, or planning, and provide limited supervision for reasoning over realistic long-tail driving scenes. We introduce nuReasoning, a large-scale real-world dataset and benchmark for reasoning-centric AD. Following the lineage of nuScenes and nuPlan, nuReasoning advances real-world AD datasets and benchmarks toward reasoning in long-tail driving scenarios. The dataset contains 20,000 clips, each 20 seconds long, collected across multiple cities, with synchronized multi-camera images, LiDAR data, HD maps, object annotations, and human-verified reasoning annotations spanning Spatial Reasoning, Decision Reasoning, and Counterfactual Reasoning. Unlike prior datasets that focus primarily on visual question answering, nuReasoning supports both reasoning evaluation and planning evaluation, enabling a direct study of how reasoning supervision affects driving performance. Experiments show that fine-tuning VLMs on nuReasoning substantially improves driving-specific question answering, while incorporating reasoning supervision into VLA training improves planning performance even when textual reasoning outputs are disabled at inference time. These results establish nuReasoning as a foundation for evaluating and improving robust, interpretable, reasoning-driven AD systems in realistic long-tail settings.

---

### What Gets Unmasked First? Trajectory Analysis of Diffusion Models for Graph-to-Text Generation

Authors: Qing Wang, Jacob Devasier, Chengkai Li
Published: 2026-05-29
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2605.31564v1

We present the first systematic study of masked diffusion language models (MDLMs) for graph-to-text generation. We analyze MDLM generation trajectories -- the order in which tokens are unmasked during iterative decoding -- and find that, unlike autoregressive LLMs which generate text linearly, MDLMs naturally prioritize entities first, followed by relational and function words, with structural tokens resolved last. We further identify a previously undocumented failure mode of supervised fine-tuning: SFT disrupts this strategy by prematurely anchoring structural sentence-ending tokens early in the decoding trajectory, effectively fixing the output length which can lead to omitted or hallucinated information. To address this, we propose lambda-scaled structural decoding, a training-free inference-time modification that downweights structural token confidence and recovers +9.4 BLEU-4. Finally, we introduce Graph-LLaDA, which integrates a Graph Transformer encoder into LLaDA's decoding process to explicitly incorporate relational graph structure. Cross-dataset evaluation on LAGRANGE reveals that previous baselines overfit to dataset-specific patterns, while LLM- and MDLM-based approaches generalize significantly better.

---

### Disagreeing Rationales: Rethinking Classification and Explainability Evaluation in Hate Speech Detection

Authors: Benedetta Muscato, Beiduo Chen, Gizem Gezici, Barbara Plank, Fosca Giannotti
Published: 2026-05-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2605.31563v1

Human disagreement is ubiquitous and well-known in labeling. However, variation in explanations, captured through token-level human rationales, remains far less explored. At the same time, it is unclear how to best evaluate human labels and rationales -- or even how to best aggregate rationales beyond majority vote -- in light of this variation. Yet, rationales may provide additional insights into the richness of human reasoning, that may differ in style, values and interpretations -- especially in subjective NLP tasks like hate speech detection. In this work, we unify diverse models, training strategies, loss functions, and existing evaluation metrics under a single protocol by systematically re-implementing them across different label and rationale representation spaces. Classification metrics are organized around two key properties -- predictive and distributional -- while explainability metrics through three complementary dimensions: plausibility, faithfulness, and complexity. In this unified supervision framework, we evaluate model behavior across classification and explainability metrics, as well as metric sensitivity to the choice of label (hard and soft) and rationale representation space (hard, intermediate and soft). Results show that both hard and soft metrics favor softer representations, highlighting their effectiveness in capturing variation and the need to rethink evaluation in subjective NLP.

---

### Effective Biological Representation Learning by Masking Gene Expression

Authors: Kian Kenyon-Dean, Alina Selega, Ihab Bendidi, Jordan M. Sorokin, Luca Bertinetto, David Errington, Hayley Donnella, Oren Kraus
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.31562v2

RNA sequencing produces rich and diverse datasets of gene expression, offering compelling insights into cellular state and function that have many applications in drug discovery. Modeling such data is challenging due to inherent technical noise and experimental batch effects, as evidenced by many existing transcriptomic foundation models (FMs) underperforming relative to linear baselines. Such results raise the question of whether deep representation learning provides a distinct advantage over the direct use of raw transcript counts. Our work explores this by developing a new self-supervised model, TxFM, with a focus on inductive representation learning evaluations. TxFM employs a masked autoencoding approach tailored to diverse RNA-seq count data, and our ablation study empirically identifies crucial architecture configurations required for strong transfer performance. Additionally, we curate a public training corpus, DiverseRNA-1.4M, and find that TxFM trained on this curated dataset yields high-fidelity gene representations that outperform FMs trained on atlas-scale corpora over 100x larger. Overall, our results indicate that inductive self-supervised learning is a viable modeling approach for transcriptomics representation, provided a careful synthesis of model architecture and training data curation.

---

### Effective Biological Representation Learning by Masking Gene Expression

Authors: Kian Kenyon-Dean, Alina Selega, Ihab Bendidi, Jordan M. Sorokin, Luca Bertinetto, David Errington, Hayley Donnella, Oren Kraus
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.31562v1

RNA sequencing produces rich and diverse datasets of gene expression, offering compelling insights into cellular state and function that have many applications in drug discovery. Modeling such data is challenging due to inherent technical noise and experimental batch effects, as evidenced by many existing transcriptomic foundation models (FMs) underperforming relative to linear baselines. Such results raise the question of whether deep representation learning provides a distinct advantage over the direct use of raw transcript counts. Our work explores this by developing a new self-supervised model, TxFM, with a focus on inductive representation learning evaluations. TxFM employs a masked autoencoding approach tailored to diverse RNA-seq count data, and our ablation study empirically identifies crucial architecture configurations required for strong transfer performance. Additionally, we curate a public training corpus, DiverseRNA-1.4M, and find that TxFM trained on this curated dataset yields high-fidelity gene representations that outperform FMs trained on atlas-scale corpora over 100x larger. Overall, our results indicate that inductive self-supervised learning is a viable modeling approach for transcriptomics representation, provided a careful synthesis of model architecture and training data curation.

---

### What Am I Missing? Question-Answering as Hidden State Probing

Authors: Chu Fei Luo, Samuel Dahan, Xiaodan Zhu
Published: 2026-05-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2605.31561v1

Test-time reasoning has become a significant field of study since the introduction of chain-of-thought reasoning in large language models (LLMs). However, the mechanisms of this reasoning process are still under-explored -- from the same input prompt, and even the same partial solution, LLMs can produce varied answers if sampled multiple times. We propose to leverage question-asking as an inference-time intervention that articulates information about the model's hidden state. To achieve that, we present a student-teacher setting where a student asks questions to a teacher. We train a probe on the student's hidden state before and after asking a question and find it is predictive of the trajectory's final correctness, even before generating the teacher's answer. This suggests there is a meaningful signal from the self-diagnosis that occurs during question generation rather than information transfer from the teacher. We then frame question-asking as a sequential decision problem, using this probe as a quality score, and define a gating policy to ask questions that maximize likelihood of correctness. We find that the success of question-asking as an intervention is largely dependent on the model's self-consistency. Our empirical results show a gap between detection and recovery; while our gating policy captures model correctness and uncertainty, interventions are equally likely to harm correct trajectories as they are to recover incorrect ones. This gap between diagnosis and correction has broader implications on language models' capacity for self-refinement under uncertainty.

---

### Functional Attention: From Pairwise Affinities to Functional Correspondences

Authors: Jiefang Xiao, Maolin Gao, Simon Weber, Guandao Yang, Daniel Cremers
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.31559v1

Learning mappings between infinite-dimensional function spaces, or operator learning, is essential for many machine learning applications. Although transformer-based operators are popular, they often rely on token-wise attention. These methods treat continuous fields as discrete tokens and usually ignore the global functional structure. We introduce \emph{Functional Attention}, which reinterprets attention as a functional correspondence between adaptive bases. Inspired by geometric functional maps, our method replaces softmax affinities with structured linear operators. This yields a compact, generalizable, resolution-invariant representation that explicitly captures global dependencies. Experiments demonstrate that \emph{Functional Attention} can match state-of-the-art performance in many operator learning tasks, including solving PDEs, 3D segmentation, and regression, while remaining robust to varying discretizations. Project page is available at https://github.com/xjffff/FUNCATTN.

---

### Positional versus Symbolic Attention Heads: Learning Dynamics, RoPE Geometry, and Length Generalization

Authors: Felipe Urrutia, Juan José Alegría, Cinthia Sanchez Macias, Jorge Salas, Cristian B. Calderon, Cristobal Rojas
Published: 2026-05-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2605.31558v1

Transformer-based language models are widespread in today's society. As such, understanding the mechanisms by which they solve structured tasks and predicting how they may behave in novel scenarios is of great importance for safe deployment. We study the learning dynamics of attention heads in a controlled setting by training a decoder-only Transformer (GPT-J) on two structurally equivalent multi-hop reasoning tasks: a number task requiring positional reasoning and a letter task requiring symbolic reasoning. Using a recently introduced metric that classifies attention-head behavior as positional or symbolic for a given prompt, we show that successful learning is associated with the emergence of pure heads, i.e., heads that express themselves as either positional or symbolic. Despite the tasks' structural equivalence, they impose different mechanistic demands: the number task requires both positional and symbolic heads, whereas the letter task requires only symbolic heads. We then identify the computational roles of these heads, characterize the basic functions they implement, and give theoretical constructions showing how single-layer RoPE-based attention can realize these functions through geometrically interpretable query, key, and value operations. This analysis yields a quantitative separation between positional and symbolic mechanisms in their robustness to longer sequences, formalized through a novel notion of discrepancy. We empirically validate the resulting predictions in both controlled and real-world models, showing that symbolic mechanisms extrapolate more reliably to longer sequences while positional mechanisms face sharper limitations.

---

### EGOSTREAM: A Diagnostic Benchmark for Streaming Episodic Memory in Egocentric Vision

Authors: Rosario Forte, Giuseppe Lando, Antonino Furnari
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31557v2

Continuous episodic memory is a core capability for autonomous agents operating in dynamic, real-world environments, yet current streaming video benchmarks provide limited tools for diagnosing what models remember and for how long. We introduce Egostream, a diagnostic benchmark for streaming episodic memory evaluation in egocentric vision. \egostream organizes 2,250 curated questions along seven cognitive dimensions: detail, spatial, temporal, event, social, causal, and prospective memory. We introduce the Answer Validity Window (AVW), which specifies the temporal span an answer remains valid as the observed scene evolves. This allows us to expand the questions into 8,528 recall-conditioned evaluations, enabling controlled testing from instant to ultra-long-term recall while separating genuine model forgetting from natural world-state changes. We rigorously establish baseline performance through a unified streaming MLLM framework that compares several state-of-the-art memory-management mechanisms, covering sliding windows, attention sinks, KV-cache pruning, merging, and offloading. Experiments within a unified Qwen3-VL backbone reveal that comparable aggregate accuracies mask starkly different memory profiles. For instance, token pruning preserves fine-grained details and temporal structure significantly better than token merging, while quantized offloading rescues ultra-long-term recall. Ultimately, all mechanisms operate well below real-time (>1s per frame), and top performing methods ceil at about 45% accuracy, exposing critical gaps in current architectures. Egostream provides the diagnostic testbed needed to close these gaps. Project website, news and updates at: https://saroo25.github.io/Egostream/

---

### Vision-Language Models Suppress Female Representations Under Ambiguous Input

Authors: Arnau Marin-Llobet, Simon Henniger, Mahzarin R. Banaji
Published: 2026-05-29
Categories: cs.CV, cs.AI, cs.CL, cs.CY, cs.HC
arXiv: https://arxiv.org/abs/2605.31556v2

Alignment teaches vision-language models (VLMs) to avoid expressing demographic biases, and when gender is clearly visible they largely succeed. Far less is known about ambiguous inputs (a worker in full gear, a figure seen from behind), cases common in practice yet rarely studied. We find that minimal prompting pressure exposes occupation-gender defaults when prompting ambiguous input images, with models collapsing to male even for strongly female-stereotyped occupations. But do these outputs reflect what models actually encode internally? We introduce LALS (Latent Association Leaning Score), a zero-shot metric that projects visual-token activations into the model's text-embedding space to measure concept associations per token and layer. Across 15 occupations, over 800 gender-ambiguous images, and four VLMs, internal representations and outputs often become systematically decoupled: models often encode a female association internally yet output male. Layer-wise analysis reveals an asymmetric filter: male signal amplifies end-to-end while female signal peaks mid-network and is suppressed before generation. A color ablation shows that culturally loaded visual cues such as clothing color further modulate these internal associations.

---

### Vision-Language Models Suppress Female Representations Under Ambiguous Input

Authors: Arnau Marin-Llobet, Simon Henniger, Mahzarin R. Banaji
Published: 2026-05-29
Categories: cs.CV, cs.AI, cs.CL, cs.CY, cs.HC
arXiv: https://arxiv.org/abs/2605.31556v1

Alignment teaches vision-language models (VLMs) to avoid expressing demographic biases, and when gender is clearly visible they largely succeed. Far less is known about ambiguous inputs (a worker in full gear, a figure seen from behind) cases common in practice yet rarely studied. We find that minimal prompting pressure exposes occupation-gender defaults when prompting ambiguous input images, with models collapsing to male even for strongly female-stereotyped occupations. But do these outputs reflect what models actually encode internally? We introduce LALS (Latent Association Leaning Score), a zero-shot metric that projects visual-token activations into the model's text-embedding space to measure concept associations per token and layer. Across 15 occupations, over 800 gender-ambiguous images, and four VLMs, internal representations and outputs are systematically decoupled: models often encode a female association internally yet output male. Layer-wise analysis reveals an asymmetric filter -- male signal amplifies end-to-end while female signal peaks mid-network and is suppressed before generation -- and a color ablation shows that culturally loaded visual cues such as clothing color further modulate these internal associations.

---

### SMART: SMPLest-X Mesh Adaptation and RAFT Tracking for Soccer Pose Estimation

Authors: Parthsarthi Rawat
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31551v1

We present our approach to the FIFA Skeletal Tracking Challenge 2026, which requires estimating 3D world-space poses of soccer players from broadcast video. Our method finetunes SMPLest-X (ViT-H, 687 M parameters) via a stratified clip split, multi-task depth supervision, and broadcast augmentation, paired with a RAFT dense optical flow camera tracker, foot-plane anchoring, and two-pass temporal smoothing. Against the FIFA baseline score of 1.053 on the validation set, SMART achieves 0.647, a 38.6% improvement; on the held-out test set, SMART scores 0.593 (Global MPJPE: 0.324 m, Local MPJPE: 0.054 m).

---

### Semantic Triplet Restoration: A Novel Protocol for Hierarchical Table Understanding in Large Language Models

Authors: Yibin Zhao, Fangxin Shang, Dingrui Yang, Yuqi Wang
Published: 2026-05-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2605.31550v1

Table question answering requires models to recover semantic relations encoded implicitly by two-dimensional layout, merged cells, and hierarchical headers. Current pipelines typically use HTML or Markdown as intermediate table representations, but these layout-oriented serializations introduce markup overhead and require large language models to infer header-cell alignments from row and column spans. We propose Semantic Triplet Restoration (STR), a protocol that rewrites each cell as an atomic fact <item path, feature path, value>, where the item path specifies the row-wise entity, the feature path specifies the hierarchical attribute, and the value contains the cell content. We also present TripletQL, a lightweight query-aware router that uses STR to select an appropriate rendering or filtered subset of triplets for each question. Across four Chinese and English table-QA benchmarks, STR matches or improves upon HTML-based baselines while reducing input tokens. The relative benefit grows for smaller language models and longer table contexts, suggesting that explicit semantic representations are especially useful under constrained inference budgets. Code and data are available at https://github.com/Phoenix-ni/STR.git .

---

### The Dynamic-Probabilistic Consistency Gap in Chaotic Surrogate Modeling

Authors: Andre Herz, Matthijs Pals, Daniel Durstewitz, Georgia Koppe
Published: 2026-05-29
Categories: cs.LG, math.DS, stat.ML
arXiv: https://arxiv.org/abs/2605.31547v1

Dynamical systems reconstruction (DSR) aims to learn surrogate models that capture the dynamics underlying time-series data. Reliably deploying these surrogates requires uncertainty estimates consistent with the learned dynamics. We expose a dynamic-probabilistic consistency (DPC) gap: the pursuit of finite-horizon probabilistic objectives can degrade dynamics or decouple predictive uncertainty from the local tangent dynamics it ought to reflect. We isolate three mechanisms behind this gap: core collapse, noise masking, and blind uncertainty. Specifically, we show that open-loop Gaussian rollout objectives can penalize Jacobian-generated covariance growth in chaotic systems, encouraging optimization shortcuts that weaken physical expansion or decouple uncertainty from it. To mitigate this gap, we propose KAFFEE (Kalman-Aware Framework For Ergodic Emulation), a differentiable extended Kalman filter-based training framework that evaluates likelihood on local predictive residuals (innovations) while transporting covariance through learned local Jacobians. On stochastic hyperchaotic Lorenz-96, KAFFEE reduces the identified failure modes, improves reconstruction of dynamical invariants relative to open-loop objectives, and maintains competitive predictive scores. We further show that the DPC gap appears when probabilistically adapting a DSR foundation model across 13 chaotic systems, where KAFFEE enables in-context Bayesian filtering while largely preserving zero-shot dynamics.

---

### Preference-Aware Rubric Learning for Personalized Evaluation

Authors: Yilun Qiu, Xiaoyan Zhao, Yang Zhang, Yuxin Chen, Cilin Yan, Jiayin Cai, Xiaolong Jiang, Yao Hu, Yoko Yamakata, Tat-Seng Chua
Published: 2026-05-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2605.31545v1

As Large Language Models (LLMs) evolve from general-purpose assistants to user-centric agents, personalization has become central to aligning model behavior with individual preferences, making the evaluation of personalized alignment a critical bottleneck. Existing evaluation methods-ranging from automatic metrics to LLM-as-a-judge approaches-fail to capture subjective, user-specific preferences embedded in long-term interaction histories. We identify three essential principles for reliable and effective personalized evaluation: Representativeness, User-Consistency, and Discriminativeness. To address these principles, we introduce Personalized Evaluation as Learning, a paradigm that formulates personalized evaluation as a learning problem rather than a static judgment. Under this paradigm, we propose PARL (Preference-Aware Rubric Learning for Personalized Evaluation), a framework that learns to induce preference-aware evaluation rubrics directly from raw user histories and performs a self-validation mechanism to ensure consistency with the user's preferences. PARL integrates rubric induction with a discriminative reinforcement learning objective that contrasts user-authored responses against competitive personalized model outputs, enabling the learned rubrics to capture precise, user-specific decision boundaries. Experiments on real-world personalized text generation tasks show that PARL consistently induces high-fidelity rubrics that reliably identify user-aligned responses and generalize across users and tasks, while capturing stable stylistic preferences and fine-grained evaluative patterns. To ensure reproducibility, our code is available at https://github.com/SnowCharmQ/PARL.

---

### Automated Prediction of Postoperative Pancreatic Fistula Using Preoperative Computed Tomography

Authors: Ashok Choudhary, Chris Varghese, Leo Y. Li-Han, Frank G. Lee, Ellen L. Larson, Elizabeth B. Habermann, Cornelius A. Thiels, Hojjat Salehinejad
Published: 2026-05-29
Categories: cs.CV, cs.LG, q-bio.QM
arXiv: https://arxiv.org/abs/2605.31539v1

Postoperative pancreatic fistula (POPF) is a serious complication after pancreatic resection, increasing morbidity, hospital stay, and healthcare costs. We present an automatic, end-to-end deep learning pipeline-from pancreatic segmentation to classification-for preoperative POPF risk estimation and stratification using preoperative CT scans. A data set with auto-segmented pancreas volumes and surgical outcomes was used to evaluate multiple architectures, including a custom lightweight 3D CNN baseline (CNN3D), R(2+1)D ResNet-18, and ResNet-MC3-18 models. Evaluation across multiple 3D architectures demonstrated promising predictive performance. This approach offers a clinically valuable tool and a methodological benchmark for pancreas-specific CT classification, supporting improved preoperative decision-making in pancreatic surgery.

---

### RayDer: Scalable Self-Supervised Novel View Synthesis from Real-World Video

Authors: Ulrich Prestel, Stefan Andreas Baumann, Nick Stracke, Björn Ommer
Published: 2026-05-29
Categories: cs.CV, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2605.31535v1

Self-supervised novel view synthesis (NVS) remains challenging to scale, despite the abundance of video data, largely due to the brittleness of training on realistic videos and the hard-to-predict scaling behavior of multi-network system designs. We introduce RayDer, a unified, feed-forward transformer that consolidates camera estimation, scene reconstruction, and rendering into a single backbone, turning self-supervised NVS into a well-posed single-model scaling problem. A minimal dynamic state, treated as a nuisance factor, absorbs time-varying content and enables stable training on unconstrained real-world video. Importantly, RayDer keeps static-scene NVS as its target task: dynamic content is leveraged purely as scalable supervision, not reconstructed as in dynamic-scene (4D) NVS. Across multiple model sizes and orders of magnitude in data, RayDer exhibits clean power-law scaling with data and compute, and outperforms static-scene data mixtures. On a large number of benchmarks, RayDer achieves strong zero-shot open-set performance competitive with state-of-the-art supervised approaches. Project Page: https://compvis.github.io/rayder

---

### Feature-Optimized Vision for Adaptive 3D Scene Reconstruction

Authors: Eric Liang
Published: 2026-05-29
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2605.31534v1

Three-dimensional scene reconstruction depends on local image evidence that is both visually discriminative and geometrically useful. Fixed feature thresholds and uniform feature budgets are easy to deploy, but they can waste computation on repeated texture, low-parallax regions, or unstable points. This paper proposes an adaptive feature-optimized vision front end for 3D reconstruction. The method scores candidate features by texture, repeatability, distinctiveness, expected triangulation angle, and spatial coverage, then allocates a per-view feature budget to maximize useful tracks under a fixed reconstruction pipeline. A small synthetic multi-view prototype evaluates four selection policies across corridor, facade, object-table, and cluttered scenes. Compared with random, texture-only, and uniform-grid baselines, the adaptive policy obtains the best quality-aware completeness and the lowest aggregate reconstruction RMSE while preserving broad image coverage. The result is not a replacement for modern learned matching or neural reconstruction systems; it is a modular front-end policy that can make classical and learned 3D pipelines more deliberate about which visual evidence they spend compute on.

---

### Discovering Thermodynamically Admissible Dissipation Potentials via Grammar-Based Symbolic Regression

Authors: Federico Califano, Jacopo Ciambella
Published: 2026-05-29
Categories: cond-mat.soft, cs.LG
arXiv: https://arxiv.org/abs/2605.31532v1

Constitutive laws for inelastic materials must satisfy strict thermodynamic admissibility requirements, yet current data-driven approaches sacrifice interpretability, even when formal guarantees are provided by physics-encoded architectures. We propose a symbolic regression framework for the data-driven discovery of dissipation potentials governing the evolution of internal variables within the Generalized Standard Materials (GSM) formalism. Starting from the Clausius--Duhem inequality, we enforce the thermodynamic requirements, convexity and non-negativity, that the dual dissipation potential must satisfy to guarantee non-negative mechanical dissipation. These requirements are formulated in the general subdifferential setting, encompassing rate-dependent (viscoelastic) and viscoplastic dissipative mechanisms, including potentials with genuine elastic domains, within a unified framework. Candidate potentials are generated by a composition-extended convexity-preserving grammar that guarantees thermodynamic admissibility \emph{by construction}. The framework is validated on synthetic datasets spanning Newtonian, power-law, and Bingham viscoplastic ground truths under process and measurement noise, and on experimental oscillatory shear measurements of a synthetic elastomer across multiple strain amplitudes and frequencies, where the discovered potentials reproduce the amplitude-dependent softening of the dynamic moduli and outperform a calibrated linear Zener baseline.

---

### SVI-Bench: A Dynamic Microworld for Strategic Video Intelligence

Authors: Yulu Pan, Han Yi, Seongsu Ha, Md Mohaiminul Islam, Benjamin Zhang, Lorenzo Torresani, Gedas Bertasius
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31529v2

True video intelligence demands more than recognizing what is visible: it requires reasoning about why events unfold, predicting what would change under different conditions, and deciding what to do next. We refer to this progression, from perception through causal reasoning and simulation to strategic planning, as Strategic Video Intelligence (SVI). No existing benchmark evaluates this capability stack: in-the-wild videos lack verifiable ground truth for causal and strategic questions, while synthetic environments sacrifice the complexity of real multi-agent systems. To bridge this gap, we introduce SVI-Bench, a large-scale benchmark that leverages team sports as a dynamic microworld, combining the complexity of real-world multi-agent interaction (10-22 agents making coordinated decisions under adversarial pressure) with the verifiability of explicit rules and definitive outcomes. SVI-Bench comprises approximately 35K hours of broadcast video, 15M annotated actions, 15K hours of expert commentary, 23K game reports, and 103K structured statistical records across basketball, soccer, and hockey, all constructed via a data engine that transforms raw game data into a dense, cross-referenced corpus. We organize evaluation into 9 tasks spanning a progressive four-pillar hierarchy: Dynamic Scene Understanding, Causal Reasoning, Strategic Simulation, and Agentic Synthesis. Evaluating strong multimodal and agentic baselines, we find a capability cliff: models perform competently on perceptual tasks, achieving approximately 74% on fine-grained action QA, but degrade sharply at each successive cognitive level. Agentic tasks prove hardest: the strongest model achieves only 5% accuracy when required to autonomously gather and integrate evidence across a corpus of 1.8M clips.

---

### Value Functions as Supermartingale Certificates

Authors: Alessandro Abate, Daniel Contro, Mirco Giacobbe, Agustín Martínez-Suñé, Diptarko Roy
Published: 2026-05-29
Categories: cs.LG, cs.LO
arXiv: https://arxiv.org/abs/2605.31524v1

Certification methods for stochastic systems provide sufficient proof rules, based on real-valued supermartingale certificates, to determine the almost-sure satisfaction of $ω$-regular properties (and therefore of linear temporal logic) over general state spaces, encompassing both countably infinite and continuous state spaces. Conversely, reinforcement learning (RL) methods for $ω$-regular tasks have received considerable attention, but they typically lack formal guarantees that the learned policy satisfies the specification, except possibly for finite state and action spaces. We bridge these two lines of research by establishing a novel theoretical connection: under an appropriate reward, the value function associated to a policy that almost surely satisfies an $ω$-regular property encodes a Streett supermartingale certificate for that specification. Our results, validated experimentally on finite Markov decision processes, hold for finite, countably infinite, and continuous state spaces, suggesting a principled route to certificate synthesis via RL.

---

### Chem-PerturBridge: a harmonized compendium of small molecule perturbation transcriptomic effects

Authors: Artur Szałata, Olga Novitskaia, Maiia Shulman, Matthew Mella, Altynbek Zhubanchaliyev, Fabian J. Theis
Published: 2026-05-29
Categories: cs.LG, q-bio.GN, q-bio.QM
arXiv: https://arxiv.org/abs/2605.31522v1

Large perturbation models require training data encompassing chemical, cellular, and assay diversity. Current transcriptomic resources for small-molecule modeling, however, are fragmented across technologies, metadata conventions, controls, doses, and preprocessing pipelines. We introduce Chem-PerturBridge, a harmonized multi-dataset resource comprising over 37k compounds, 136 cellular contexts, and 1.25M transcriptomic samples across eight assay types, with standardized identifiers, metadata, and replicate-aware condition-level effects. We use the resource to evaluate matched-condition agreement across datasets and replicate agreement within datasets. Matched same-compound conditions generally show weak agreement in fine-grained logFC rankings and magnitudes across most dataset pairs, often falling below same-context different-compound baselines. In contrast, logFC direction agreement is substantially more stable and usually exceeds these baselines. We further evaluate Chem-PerturBridge as a pretraining resource for compound representation learning. Under a compound-held-out OP3 evaluation split, embeddings pretrained on Chem-PerturBridge improve over L1000-only embeddings, Morgan fingerprints, and the descriptor-free OP3 baseline across metrics. An extensive molecule-holdout evaluation across 11 datasets further shows that models trained on Chem-PerturBridge outperform or match those that are not. Chem-PerturBridge therefore supports both diagnostic evaluation of cross-dataset signature agreement and model-oriented reuse of heterogeneous perturbation transcriptomic data.

---

### UniAudio-Token: Empowering Semantic Speech Tokenizers with General Audio Perception

Authors: Yuhan Song, Linhao Zhang, Aiwei Liu, Chuhan Wu, Sijun Zhang, Wei Jia, Yuan Liu, Houfeng Wang, Xiao Zhou
Published: 2026-05-29
Categories: cs.CL, cs.SD
arXiv: https://arxiv.org/abs/2605.31521v1

Semantic speech tokenizers have become a widely used interface for Audio-LLMs, owing to their compact single-codebook design and strong linguistic alignment. However, their focus on linguistic abstraction induces acoustic blindness, limiting their applicability beyond speech-centric tasks. We propose UniAudio-Token, a framework that empowers semantic tokenizers with general audio perception without compromising speech ability. Instead of altering the semantic paradigm, UniAudio-Token mitigates its information loss through two key innovations: (1) Semantic-Acoustic Primitives (SAP) provide structured supervision by decomposing audio into linguistic content, vocal attributes, and auditory-scene primitives; and (2) Semantic-Acoustic Equilibrium (SAE) introduces a content-aware gating mechanism that adaptively restores fine-grained acoustic details from shallow layers. Extensive evaluations show that UniAudio-Token learns comprehensive universal representations while preserving high-fidelity speech generation. When integrated with downstream LLMs, it outperforms all single-codebook baseline tokenizers on both understanding and generation tasks, effectively serving as a unified audio interface. We publicly release all our code, including training and inference scripts, together with the model checkpoints at https://github.com/Tencent/Universal_Audio_Tokenizer.

---

### Separating Secrets from Placeholders: A Hybrid CNN-CodeBERT Framework for Three-Class Credential Leakage Detection

Authors: Maksuda Bilkis Baby, Khushika Shah, Naiyue Liang, Lei Zhang
Published: 2026-05-29
Categories: cs.SE, cs.AI, cs.CR
arXiv: https://arxiv.org/abs/2605.31520v1

Credential leakage in public source code repositories poses a critical security threat, with over 23.8 million secrets exposed in 2024 alone. Existing detection tools suffer from high false-positive rates because rigid pattern matching and binary classification schemes fail to distinguish genuine credentials from placeholder or weak credentials. We propose a three-class classification framework that explicitly models placeholder or weak credentials as a distinct class, leveraging CodeBERT-based semantic understanding combined with character-level pattern recognition. We evaluate our approach on a newly constructed dataset of 9,426 samples spanning 10 programming languages. Our model achieves a Matthews Correlation Coefficient of 0.86 and a macro F1-score of 0.90, achieving 93% recall and 89% precision for genuine credential leaks while reducing high severity alerts by 33.0% (from 373 to 250) without sacrificing security coverage. Compared to prior character-level approaches, our method improves placeholder or weak credential detection from 54% to 81% F1-score while maintaining strong cross language generalization, with 9 of 10 languages achieving F1 above 0.80 under leave-one-language-out evaluation.

---

### On the Relationship Between Activation Outliers and Feature Death in Sparse Autoencoders

Authors: Elana Simon, Etowah Adams, James Zou
Published: 2026-05-29
Categories: cs.LG
arXiv: https://arxiv.org/abs/2605.31518v1

Sparse autoencoders (SAEs) decompose neural network activations into interpretable features, but many learned features never activate, a problem called feature death that wastes dictionary capacity and can reintroduce superposition. Death rates vary dramatically between models: near-zero on GPT-2, over 70% on AlphaFold3 with identical configurations. We find that dimension-level activation outliers (dimensions whose mean magnitude is large relative to per-token variation) cause this by shifting pre-activations at initialization based on each feature's alignment with the activation mean. Features anti-aligned with the mean receive permanently negative pre-activations and never fire. We formalize outlier severity as $γ= \|μ\|/\|σ\|$; it predicts initial death rates (Spearman $ρ= 0.89$ for dead-by-TopK, $0.82$ for dead-by-ReLU) across 454 model-layer combinations spanning language, vision, protein, and genomic models. Dead features can revive during training, but recovery requires the SAE bias to learn the activation mean, a process that is prohibitively slow at high $γ$. Mean-centering (subtracting the activation mean) sidesteps this and eliminates outlier-induced death across all tested models, confirming the mechanism and providing a principled basis for when and why this preprocessing step is necessary.

---

### If LLMs Have Human-Like Attributes, Then So Does Age of Empires II

Authors: Adrian de Wynter
Published: 2026-05-29
Categories: cs.CL, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2605.31514v3

Much research has been carried out on large language models (LLMs) and LLM-powered agentic workflows. However, many works within the field state emergence of, ascribe to, or assume, generalised anthropomorphic attributes to them (e.g., morality or understanding of natural language). Our goal is not to argue in favour or against the existence of these attributes, but to point out that these conclusions could be incorrect. For this we build and train a simple neural network on the videogame Age of Empires II, and note that any entity in a sufficiently-powerful substrate, such as LEGO or the Greater Boston Area, could also present such attributes. Hence, the purported anthropomorphic attributes of LLMs are empirically non-unique: although some properties (e.g., responses to prompts) could remain invariant, others, such as the interpretation of their perceived behaviour, might change with the substrate. Thus, any empirically-grounded discussion on these attributes requires explicit measurement criteria; otherwise the interpretation is left to the representation. We then show that assuming that these attributes exist or not in a system, independent of the substrate and in a generalised way, leads to either circular or uninformative conclusions. This is regardless of the experimenter's viewpoint on the subject, or whether the outcome shows existence or non-existence. Finally we propose a 'null' assumption, where one assumes LLM non-uniqueness instead of assuming anthropomorphic attributes to set up an experiment, along with examples of it. We also discuss potential objections to our work, briefly survey the field, and prove that Age of Empires II is functionally- and Turing-complete.

---

### Personalize Your Large Vision-language Models With In-context Prompt Tuning

Authors: Yanshu Li, Jiaqian Li, Kuai Yu, Xi Xiao, Dongfang Liu, Tianyang Wang, Ruixiang Tang
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31513v2

Large vision-language models (LVLMs) have demonstrated strong general multimodal capability and are increasingly deployed in downstream systems. This trend has driven growing interest in LVLM personalization, which aims to enable models to quickly and effectively learn out-of-distribution multimodal concepts to meet user-specific needs. However, many existing methods rely on inference-time training, which reduces efficiency. They also struggle to maintain accuracy in complex multi-image, multi-concept settings. These limitations restrict the broader deployment of LVLM-based systems. Therefore, this paper proposes in-context prompt tuning (ICPT). Specifically, ICPT employs a lightweight projection module capable of operating in complex scenarios to extract fine-grained visual semantics from multiple reference images, seamlessly transforming these features alongside identity-label mappings into continuous prompts. To maximize computational efficiency, this module adaptively determines the prompt length based on the intrinsic visual complexity of each concept. Crucially, to overcome the environmental biases and cross-concept interference prevalent in real-world applications, we introduce two novel geometric regularizations. These constraints refine prompt representations by decoupling key identities from transient environmental states and separating concepts to avoid semantic confusion. Extensive experiments show that ICPT achieves state-of-the-art personalization accuracy across diverse tasks and LVLM backbones.

---

### Reliable Multilingual Orthopedic Decision Support from Clinical Narratives: Language-Aware Adaptation and Verification-Guided Deferral

Authors: Danish Ali, Li Xiaojian, Sundas Iqbal, Farrukh Zaidi
Published: 2026-05-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2605.31512v1

Multilingual orthopedic decision support remains challenging in low-resource healthcare settings, where clinical narratives contain specialized terminology, mixed scripts, incomplete evidence, label imbalance and language-dependent documentation patterns. This article presents a reliability-oriented framework for classifying free-text orthopedic notes in English, Hindi and Punjabi. We compare task-aligned multilingual transformer encoders, a task-fine-tuned DistilBERT baseline, zero-shot instruction-tuned large language models (LLMs) and a domain-adaptive encoder, IndicBERT-HPA. IndicBERT-HPA augments IndicBERT with language-aware orthopedic adapter heads to support clinically relevant multilingual representation learning. Evaluation extends beyond aggregate accuracy to per-class performance, ROC-AUC, AUPRC, expected calibration error, cross-language stability and robustness under controlled balanced and natural-prevalence distributions. The evaluated zero-shot LLMs remain substantially less effective than task-adapted encoders for closed-set classification, with language-dependent instability. Under natural clinical prevalence, IndicBERT-HPA achieves the strongest overall performance, reaching an averaged Macro-F1 of 0.8792, Macro-AUROC of 0.894 and AUPRC of 0.902. We further implement a deterministic selective-verification layer combining confidence gating, evidence-consistency checking and language-risk screening. On a randomly selected held-out 5,000-record subset, it achieves 84.4% selective accuracy and 0.76 selective Macro-F1 at 72.3% coverage, compared with 71.5% accuracy and 0.65 Macro-F1 for accept-all prediction. These results support reliability-oriented multilingual clinical decision support with explicit deferral.

---

### Skill Reuse as Compression in Agentic RL

Authors: Zhikun Xu, Yu Feng, Jacob Dineen, Taiwei Shi, Jieyu Zhao, Ben Zhou
Published: 2026-05-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2605.31509v2

Large language model agents trained with reinforcement learning (RL) often learn brittle, task-specific shortcuts. We hypothesize that agents generalize better when their successful trajectories are structurally compressible, decomposed into a small set of reusable abstract patterns. To formalize this, we introduce ReuseRL, which grounds agentic RL in the Minimum Description Length (MDL) principle. ReuseRL extracts a shared skill dictionary from successful trajectories and augments the RL objective with a segmentation cost, explicitly penalizing idiosyncratic behaviors that encode poorly. We prove a PAC-Bayes bound guaranteeing that a dictionary extracted from successful trajectories has bounded expected description length on future successful behavior. Across ALFWorld, TextWorld-Cooking, and Countdown-Stepwise, ReuseRL improves in- and out-of-distribution success over vanilla GRPO and strong round-length baselines.

---

### Skill Reuse as Compression in Agentic RL

Authors: Zhikun Xu, Yu Feng, Jacob Dineen, Taiwei Shi, Jieyu Zhao, Ben Zhou
Published: 2026-05-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2605.31509v1

Large language model agents trained with reinforcement learning (RL) often learn brittle, task-specific shortcuts. We hypothesize that agents generalize better when their successful trajectories are structurally compressible, decomposed into a small set of reusable abstract patterns. To formalize this, we introduce ReuseRL, which grounds agentic RL in the Minimum Description Length (MDL) principle. ReuseRL extracts a shared skill dictionary from successful trajectories and augments the RL objective with a segmentation cost, explicitly penalizing idiosyncratic behaviors that encode poorly. We prove a PAC-Bayes generalization bound for this compression penalty. Across ALFWorld, TextWorld-Cooking, and Countdown-Stepwise, ReuseRL improves in- and out-of-distribution success over vanilla GRPO and strong round-length baselines.

---

### Internalizing Temporal Consistency in Video Object-Centric Learning without Explicit Regularization

Authors: Rongzhen Zhao, Zhiyuan Li, Juho Kannala, Joni Pajarinen
Published: 2026-05-29
Categories: cs.CV
arXiv: https://arxiv.org/abs/2605.31508v1

Video Object-Centric Learning (OCL) aims to represent objects as \textit{slot} vectors and maintain their consistency across frames. Slot-Slot Contrastive (SSC) loss has become the cornerstone for state-of-the-art (SOTA) video OCL methods. While highly effective, SSC relies on one-to-one object correspondence across frames and introduces an extra loss. Following Occam's Razor, we propose a paradigm shift: temporal consistency is better enforced as an implicit model design rather than an explicit loss. To elegantly exclude SSC (\textbf{xSSC}), we introduce two quasi-zero-overhead synergistic mechanisms: (\textit{i}) Chrono-Channel Decomposition (CCD) structurally disentangles slot representations along the channel dimension into \textit{static} and \textit{dynamic} sub-spaces, serving as an empirically unified information bottleneck; (\textit{ii}) Cross-Temporal Reconstruction (CTR) stochastically reconstructs target features of either the current or previous time step by fusing current slots' static channels and target slots' dynamic channels, using a single standard OCL decoder with minor training adaptation. Thereby, the slot sets inherently learn temporal consistency by minimizing the standard reconstruction error alone. Extensive experiments show that integrating xSSC into leading baselines not only improves training efficiency but also establishes new SOTAs on video object discovery and recognition tasks. Furthermore, our PCA and gradient analyses confirm that objects' time-invariant semantics and time-variant kinematics are encoded into the proposed sub-spaces. Our source code, model checkpoints and training logs are provided on https://github.com/Genera1Z/xSSC.

---

### Evaluating Factual Density in Multi-Source RAG: A Study in Medical AI Accuracy

Authors: Michael R. DeMarco
Published: 2026-05-29
Categories: cs.IR, cs.CL
arXiv: https://arxiv.org/abs/2605.31506v2

Retrieval-Augmented Generation (RAG) is the current industry standard for grounding AI in real-world facts. Traditional retrieval methods rely on keyword matching and topic proximity, ranking content based on how closely it sounds like the user's query. What they do not measure is how many verified facts the content actually contains. This structural gap, termed the Expert Blindness Effect, causes standard RAG pipelines to consistently bury high-density factual evidence in favor of lexically dominant text on the same topic. To address this gap, this paper introduces Factual Density (FD*), a novel retrieval optimization signal that measures the proportion of verified atomic claims relative to total token count. Using the NexusAgentics Ghost Audit preprocessing pipeline, raw text is scored for factual specificity using probabilistic factuality analysis to filter content before corpus ingestion. An initial formulation introduced a severe document-length confound (Pearson R = -0.8636, p = 2.27e-07). Implementing Z-score normalization within length bins resolved this bias, validating FD* as a length-independent density signal (p = 0.0749). Evaluated against the HealthFC benchmark (750 health claims labeled Supported, Refuted, or No Evidence by medical experts), FD*-optimized retrieval was the only condition to achieve 100% systematic review saturation in top-5 results, surfacing Cochrane evidence that standard cosine similarity ranked outside the top ten. Ground truth verification confirmed 25 mappings across seven HealthFC-supported claims. While full statistical validation across n=50 queries remains future work due to constraints on corpus-benchmark alignment, these findings establish factual density reranking as a low-cost, high-impact intervention for improving factual precision in health RAG architectures.

---

### When Are Multimodal Predictions Biologically Supported? A Diagnostic Evaluation Framework

Authors: Dylan Steiner, Gustavo Arango-Argoty, Gerald Sun, Etai Jacob
Published: 2026-05-29
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2605.31504v1

Multimodal models in oncology can produce accurate predictions, but accurate prediction does not reveal whether the model has learned biology that is shared across modalities, biology confined to one modality, or spurious correlations that reflect confounders rather than genuine biology. We introduce DECAT, a model-agnostic post-hoc evaluation framework that classifies multimodal representations into four diagnostic scenarios for a given task and modality, using five null-referenced metrics and a rule-based decision procedure. The framework operates on learned representations, requires no knowledge of which specific confounder is present, and returns indeterminate when the evidence is insufficient. We validate DECAT on synthetic data across four multimodal model classes (over 2,500 trained representations) and on real data from 8,979 TCGA patients, evaluating both multimodal embeddings and five pretrained pathology foundation models. Entangled models (e.g., CLIP) achieve near-perfect shared biology detection but falsely claim shared biology in the majority of cases where it is absent on real foundation model embeddings. This false claim rate increases with confound strength so that larger cohorts and stronger representations produce more confident but still incorrect diagnoses. Applied to both multimodal TCGA embeddings and five pathology foundation models without paired RNA, DECAT detects confounding invisible to AUROC without requiring the confounder labels, as confirmed by post-hoc stratification.

---

### How can embedding models bind concepts?

Authors: Arnas Uselis, Darina Koishigarina, Seong Joon Oh
Published: 2026-05-29
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2605.31503v1

Humans easily determine which color belongs to which shape in multi-object scenes, an ability known as concept binding. Vision-language embedding models such as CLIP struggle with binding: they recognize individual concepts but fail to represent which concepts form which objects. Although CLIP behaves like a bag-of-concepts model in cross-modal retrieval, object information is recoverable from its image and text embeddings separately. We study this tension through the binding function, which maps concepts to scene embeddings. We find that scene embeddings decompose additively into object representations, explaining why uni-modal probes can recover object information. However, CLIP's binding function is high-complexity, which likely prevents the image and text encoders from learning a shared binding mechanism that generalizes to unseen concept combinations. We then ask whether this limitation is fundamental. We show that it is not. In controlled transformer models trained from scratch, binding generalization emerges with sufficient data coverage. These models learn low-complexity binding functions characterized by multiplicative interactions between concepts, enabling systematic generalization. Code is publicly available at https://github.com/oshapio/binding-concepts-complexity.

---

### On Efficient Scaling of GNNs via IO-Aware Layers Implementations

Authors: Daria Fomina, Daniil Krasylnikov, Alexey Boykov, Andrey Dolgovyazov, Vyacheslav Zhdanovskiy, Fedor Velikonivtsev
Published: 2026-05-29
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2605.31500v1

Graph Neural Networks (GNNs) are bottlenecked by sparse, irregular memory access. Popular frameworks such as DGL and PyTorch Geometric support general message passing, but complex layers often materialize edge-wise intermediates, increasing memory traffic and limiting scalability on large graphs. We take an I/O- and arithmetic-intensity--centric view and show that widely used layers fall into three kernel families: SpMM-based convolutions, reduction-based aggregations, and attention-based layers (GATv2/Graph Transformer). For each family, we develop GPU kernels that reduce data movement, improve locality, and remain robust across realistic graphs. We also study graph reordering and find that its impact depends on the kernel mapping: it benefits neighbor-parallel (gather-dominated) kernels more consistently than feature-parallel designs. Empirically, our fused attention kernels reach up to $\textbf{3.9}\times$ speedup for Graph Transformer (median $\textbf{1.6}\times$), with Tensor Core (block-sparse) variants up to $\textbf{7.3}\times$ on locally dense graphs; for GATv2 we reach up to $\textbf{8.5}\times$ speedup (median $\textbf{2.0}\times$) while reducing peak memory by up to $\textbf{76}\times$ (median $\textbf{6}\times$). Our degree-aware reduction kernels achieve up to $\textbf{10}\times$ speedup (median $\textbf{2.6}\times$). For SpMM-based layers, properly cached cuSPARSE achieves up to $\textbf{8}\times$ speedup over DGL and outperforms evaluated custom baselines in the majority of evaluations. We release our implementations as drop-in replacements to support reproducible, hardware-aware GNN acceleration.

---

### Scalable Inference-Time Annealing with Surrogate Likelihood Estimators

Authors: Daniel Peñaherrera, Rishal Aggarwal, David Ryan Koes
Published: 2026-05-29
Categories: cs.LG, q-bio.BM
arXiv: https://arxiv.org/abs/2605.31498v3

A long standing challenge in computational chemistry and biophysics is efficiently sampling the Boltzmann distribution of molecules. Advances in generative modeling have been proposed to address the limitations of conventional sampling techniques by eliminating the computational cost of simulation. A promising direction is iteratively finetuning diffusion models along a temperature ladder whereby training data is generated via importance sampling during inference-time annealing. Unfortunately, these methods require computing a divergence over the score field to estimate importance weights, rendering them intractable for larger systems. Here we present scalable inference-time annealing (SITA), which retrains flow-based models to generate samples at progressively lower temperatures using an energy-based model to facilitate fast surrogate likelihoods. We demonstrate state-of-the-art performance on both Alanine Dipeptide and Alanine Tripeptide while avoiding costly divergence terms. Our code is available at https://github.com/countrsignal/sita.git

---

### Assign and Add: A Mechanistic Study of Compositional Arithmetic

Authors: Brady Exoo, Alberto Bietti, John Sous
Published: 2026-05-29
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2605.31497v1

Large language models are able to compose skills in order to perform complex tasks, many of which might not have been seen during training. The details of how exactly this composition occurs remain elusive. In this paper, we study a mechanism for compositional generalization in transformers by considering a simple controlled setting involving variable assignment and modular addition. By partitioning our training data into disjoint sets, we observe that small transformers are able to generalize to previously unseen combinations of variables and numbers. Our mechanistic analysis shows that the same ``modular addition'' MLP module is used whether the inputs are given directly or indirectly through a separate variable assignment mechanism. We also analyze the training dynamics from an empirical lens, which reveals three phases of learning: first, modular addition is learned, then the structure required for variable assignment, and finally a refinement phase where the model generalizes to some hard sequences not seen in training. Finally, we provide a theoretical framework to explain how compositionality emerges from training dynamics. These results suggest that compositional generalization can be a natural consequence of the compositionality of internal mechanisms in~transformers.

---

### Consolidating Rewarded Perturbations for LLM Post-Training

Authors: Zheyu Zhang, Shuo Yang, Gjergji Kasneci
Published: 2026-05-29
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2605.31494v1

Post-training of language models is commonly framed as a sample-score-update loop implemented by gradient descent. A recent line of work, exemplified by RandOpt, relocates this loop to weight space, sampling Gaussian perturbations around a pretrained model and ensembling the top-K rewarded specialists at inference. While competitive with PPO and GRPO under matched training compute, this prediction-level ensemble incurs K forward passes per test example and does not extend cleanly to free-form generation. We ask whether the rewarded population can instead be folded into a single deployable model, replacing the inference-time ensemble with one consolidated update. A split-half analysis over 25 model-task pairs reveals reproducible low-rank structure in every case. We turn this geometry into CoRP (Consolidating Rewarded Perturbations), a gradient-free operator that combines reward-weighted aggregation, compatibility-aware reweighting, and a held-out validation gate, with no gradient flowing through the language model. Across five language models from 0.5B to 8B and five tasks covering math, code, and creative writing, CoRP improves the base model by 8.1 points on average. Using one tenth of RandOpt's perturbation budget, CoRP exceeds single-inference RandOpt by 6.5 points and recovers more than half of the gain of the 50-pass majority-vote ensemble, at one forward pass per test example.

---

### LinTree: Improving LLM Reasoning with Explicitly Structured Search Histories

Authors: Liwei Kang, Yee Whye Teh, Wee Sun Lee
Published: 2026-05-29
Categories: cs.AI
arXiv: https://arxiv.org/abs/2605.31492v1

Large language models (LLMs) often solve reasoning problems by generating intermediate traces that explore and revise partial solutions. From a search perspective, these traces can be viewed as linearized search trees, where the model extends a partial solution, abandons it when it fails, and backtracks to try alternatives. Compared with traditional heuristic-guided search, such a policy has a potential advantage: it conditions on the whole search trace rather than only on the current local state. We first test whether LLMs utilize this advantage by comparing trace-conditioned reasoning policies against best-first search equipped with an LLM heuristic that only observes the current local state. Across three controlled reasoning environments, Blocks World, grid Navigation, and Sokoban, we find that raw access to search history alone is not enough to reliably outperform heuristic search. We then study one possible reason: in LLM reasoning traces, the underlying search tree is only implicitly represented, and when the model backtracks or switches branches, the trace does not explicitly identify which earlier search state is being revisited. We show that adding simple parent pointers to explicitly represent the linearized tree (LinTree) structure improves both task performance and search efficiency relative to implicit reasoning models and LLM-heuristic-guided search. These results suggest that search history becomes most useful when its tree structure is made explicit, motivating more structure-aware representations for LLM reasoning.

---

### Are Full Rollouts Necessary for On-Policy Distillation?

Authors: Yaocheng Zhang, Jiajun Chai, Yuqian Fu, Songjun Tu, Xiaohan Wang, Wei Lin, Guojun Yin, Qichao Zhang, Yuanheng Zhu, Dongbin Zhao
Published: 2026-05-29
Categories: cs.CL
arXiv: https://arxiv.org/abs/2605.31490v2

On-policy distillation (OPD) provides dense teacher feedback along student-generated rollouts rather than fixed teacher traces and has emerged as a promising post-training paradigm. However, standard OPD typically generates full rollouts during training, which is computationally expensive and may expose the student to unreliable teacher feedback at late rollout positions, especially during early training. We identify the rollout horizon as a key bottleneck in OPD that substantially impacts training efficiency. Unlike Reinforcement Learning with Verifiable Rewards (RLVR), OPD does not require a final answer reward to provide learning signals. Therefore, full rollouts may not always be necessary for OPD. Motivated by this insight, we propose two simple horizon-control strategies: Progressive OPD (POPD), which gradually expands the rollout horizon during training, and Truncated OPD (TOPD), which permanently performs distillation on reliable truncated rollouts. Experiments on mathematical reasoning show that POPD improves the training efficiency of OPD by up to 3$\times$, while TOPD matches OPD performance using only 10\% of the rollout horizon, leading to substantial wall-clock and memory reductions. These results demonstrate that controlling the rollout horizon offers a simple and practical path to more efficient OPD.

---