# Research Papers — 2026-03 · bundle page 234 of 243

> Full-text markdown of papers #11651–11700 of 12123 published in 2026-03, newest first. AI-readable mirrors of the full 12123-paper 2026-03 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-03/page/234
[← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/233.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-03/page/235.md)

### Uniform-in-time concentration in two-layer neural networks via transportation inequalities

Authors: Arnaud Guillin, Boris Nectoux, Paul Stos
Published: 2026-03-02
Categories: cs.NE, math.PR
arXiv: https://arxiv.org/abs/2603.01842v1

We quantify, uniformly over time and with high probability, the discrepancy between the predictions of a two-layer neural network trained by stochastic gradient descent (SGD) and their mean-field limit, for quadratic loss and ridge regularization. As a key ingredient, we establish T p transportation inequalities (p $\in$ {1, 2}) for the law of the SGD parameters, with explicit constants independent of the iteration index. We then prove uniform-in-time concentration of the empirical parameter measure around its mean-field limit in the Wasserstein distance W 1 , and we translate these bounds into prediction-error estimates against a fixed test function $Φ$. We also derive analogous concentration bounds in the sliced-Wasserstein distance SW 1 , leading to dimension-free rates.

---

### Trivial Graph Features and Classical Learning are Enough to Detect Random Anomalies

Authors: Matthieu Latapy, Stephany Rajeh
Published: 2026-03-02
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.01841v1

Detecting anomalies in link streams that represent various kinds of interactions is an important research topic with crucial applications. Because of the lack of ground truth data, proposed methods are mostly evaluated through their ability to detect randomly injected links. In contrast with most proposed methods, that rely on complex approaches raising computational and/or interpretability issues, we show here that trivial graph features and classical learning techniques are sufficient to detect such anomalies extremely well. This basic approach has very low computational costs and it leads to easily interpretable results. It also has many other desirable properties that we study through an extensive set of experiments. We conclude that detection methods should now target more complex kinds of anomalies.

---

### FireRed-OCR Technical Report

Authors: Hao Wu, Haoran Lou, Xinyue Li, Zuodong Zhong, Zhaojun Sun, Phellon Chen, Xuanhe Zhou, Kai Zuo, Yibo Chen, Xu Tang, Yao Hu, Boxiang Zhou et al.
Published: 2026-03-02
Categories: cs.CV, eess.IV
arXiv: https://arxiv.org/abs/2603.01840v1

We present FireRed-OCR, a systematic framework to specialize general VLMs into high-performance OCR models. Large Vision-Language Models (VLMs) have demonstrated impressive general capabilities but frequently suffer from ``structural hallucination'' when processing complex documents, limiting their utility in industrial OCR applications. In this paper, we introduce FireRed-OCR, a novel framework designed to transform general-purpose VLMs (based on Qwen3-VL) into pixel-precise structural document parsing experts. To address the scarcity of high-quality structured data, we construct a ``Geometry + Semantics'' Data Factory. Unlike traditional random sampling, our pipeline leverages geometric feature clustering and multi-dimensional tagging to synthesize and curate a highly balanced dataset, effectively handling long-tail layouts and rare document types. Furthermore, we propose a Three-Stage Progressive Training strategy that guides the model from pixel-level perception to logical structure generation. This curriculum includes: (1) Multi-task Pre-alignment to ground the model's understanding of document structure; (2) Specialized SFT for standardizing full-image Markdown output; and (3) Format-Constrained Group Relative Policy Optimization (GRPO), which utilizes reinforcement learning to enforce strict syntactic validity and structural integrity (e.g., table closure, formula syntax). Extensive evaluations on OmniDocBench v1.5 demonstrate that FireRed-OCR achieves state-of-the-art performance with an overall score of 92.94\%, significantly outperforming strong baselines such as DeepSeek-OCR 2 and OCRVerse across text, formula, table, and reading order metrics. We open-source our code and model weights to facilitate the ``General VLM to Specialized Structural Expert'' paradigm.

---

### LEAR: Learning Edge-Aware Representations for Event-to-LiDAR Localization

Authors: Kuangyi Chen, Jun Zhang, Yuxi Hu, Yi Zhou, Friedrich Fraundorfer
Published: 2026-03-02
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2603.01839v1

Event cameras offer high-temporal-resolution sensing that remains reliable under high-speed motion and challenging lighting, making them promising for localization from LiDAR point clouds in GPS-denied and visually degraded environments. However, aligning sparse, asynchronous events with dense LiDAR maps is fundamentally ill-posed, as direct correspondence estimation suffers from modality gaps. We propose LEAR, a dual-task learning framework that jointly estimates edge structures and dense event-depth flow fields to bridge the sensing-modality divide. Instead of treating edges as a post-hoc aid, LEAR couples them with flow estimation through a cross-modal fusion mechanism that injects modality-invariant geometric cues into the motion representation, and an iterative refinement strategy that enforces mutual consistency between the two tasks over multiple update steps. This synergy produces edge-aware, depth-aligned flow fields that enable more robust and accurate pose recovery via Perspective-n-Point (PnP) solvers. On several popular and challenging datasets, LEAR achieves superior performance over the best prior method. The source code, trained models, and demo videos are made publicly available online.

---

### Constrained Particle Seeking: Solving Diffusion Inverse Problems with Just Forward Passes

Authors: Hongkun Dou, Zike Chen, Zeyu Li, Hongjue Li, Lijun Yang, Yue Deng
Published: 2026-03-02
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2603.01837v1

Diffusion models have gained prominence as powerful generative tools for solving inverse problems due to their ability to model complex data distributions. However, existing methods typically rely on complete knowledge of the forward observation process to compute gradients for guided sampling, limiting their applicability in scenarios where such information is unavailable. In this work, we introduce \textbf{\emph{Constrained Particle Seeking (CPS)}}, a novel gradient-free approach that leverages all candidate particle information to actively search for the optimal particle while incorporating constraints aligned with high-density regions of the unconditional prior. Unlike previous methods that passively select promising candidates, CPS reformulates the inverse problem as a constrained optimization task, enabling more flexible and efficient particle seeking. We demonstrate that CPS can effectively solve both image and scientific inverse problems, achieving results comparable to gradient-based methods while significantly outperforming gradient-free alternatives. Code is available at https://github.com/deng-ai-lab/CPS.

---

### Affine Correspondences in Stereo Vision: Theory, Practice, and Limitations

Authors: Levente Hajder
Published: 2026-03-02
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.01836v1

Affine transformations have been recently used for stereo vision. They can be exploited in various computer vision application, e.g., when estimating surface normals, homographies, fundamental and essential matrices. Even full 3D reconstruction can be obtained by using affine correspondences. First, this paper overviews the fundamental statements for affine transformations and epipolar geometry. Then it is investigated how the transformation accuracy influences the quality of the 3D reconstruction. Besides, we propose novel techniques for estimating the local affine transformation from corresponding image directions; moreover, the fundamental matrix, related to the processed image pair, can also be exploited. Both synthetic and real quantitative evaluations are implemented based on the accuracy of the reconstructed surface normals. For the latter one, a special object, containing three perpendicular planes with chessboard patterns, is constructed. The quantitative evaluations are based on the accuracy of the reconstructed surface normals and it is concluded that the estimation accuracy is around a few degrees for realistic test cases. Special stereo poses and plane orientations are also evaluated in detail.

---

### Probing Materials Knowledge in LLMs: From Latent Embeddings to Reliable Predictions

Authors: Vineeth Venugopal, Soroush Mahjoubi, Elsa Olivetti
Published: 2026-03-02
Categories: cond-mat.mtrl-sci, cs.LG
arXiv: https://arxiv.org/abs/2603.01834v1

Large language models are increasingly applied to materials science, yet fundamental questions remain about their reliability and knowledge encoding. Evaluating 25 LLMs across four materials science tasks -- over 200 base and fine-tuned configurations -- we find that output modality fundamentally determines model behavior. For symbolic tasks, fine-tuning converges to consistent, verifiable answers with reduced response entropy, while for numerical tasks, fine-tuning improves prediction accuracy but models remain inconsistent across repeated inference runs, limiting their reliability as quantitative predictors. For numerical regression, we find that better performance can be obtained by extracting embeddings directly from intermediate transformer layers than from model text output, revealing an ``LLM head bottleneck,'' though this effect is property- and dataset-dependent. Finally, we present a longitudinal study of GPT model performance in materials science, tracking four models over 18 months and observing 9--43\% performance variation that poses reproducibility challenges for scientific applications.

---

### Uncertainty Quantification of Click and Conversion Estimates for the Autobidding

Authors: Ivan Zhigalskii, Andrey Pudovikov, Aleksandr Katrutsa, Egor Samosvat
Published: 2026-03-02
Categories: cs.LG, cs.GT, stat.ML
arXiv: https://arxiv.org/abs/2603.01825v1

Modern e-commerce platforms employ various auction mechanisms to allocate paid slots for a given item. To scale this approach to the millions of auctions, the platforms suggest promotion tools based on the autobidding algorithms. These algorithms typically depend on the Click-Through-Rate (CTR) and Conversion-Rate (CVR) estimates provided by a pre-trained machine learning model. However, the predictions of such models are uncertain and can significantly affect the performance of the autobidding algorithm. To address this issue, we propose the DenoiseBid method, which corrects the generated CTRs and CVRs to make the resulting bids more efficient in auctions. The underlying idea of our method is to employ a Bayesian approach and replace noisy CTR or CVR estimates with those from recovered distributions. To demonstrate the performance of the proposed approach, we perform extensive experiments on the synthetic, iPinYou, and BAT datasets. To evaluate the robustness of our approach to the noise scale, we use synthetic noise and noise estimated from the predictions of the pre-trained machine learning model.

---

### OpenAutoNLU: Open Source AutoML Library for NLU

Authors: Grigory Arshinov, Aleksandr Boriskin, Sergey Senichev, Ayaz Zaripov, Daria Galimzianova, Daniil Karpov, Leonid Sanochkin
Published: 2026-03-02
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2603.01824v1

OpenAutoNLU is an open-source automated machine learning library for natural language understanding (NLU) tasks, covering both text classification and named entity recognition (NER). Unlike existing solutions, we introduce data-aware training regime selection that requires no manual configuration from the user. The library also provides integrated data quality diagnostics, configurable out-of-distribution (OOD) detection, and large language model (LLM) features, all within a minimal lowcode API. The demo app is accessible here https://openautonlu.dev.

---

### Emerging Human-like Strategies for Semantic Memory Foraging in Large Language Models

Authors: Eric Lacosse, Mariana Duarte, Peter M. Todd, Daniel C. McNamee
Published: 2026-03-02
Categories: cs.AI
arXiv: https://arxiv.org/abs/2603.01822v1

Both humans and Large Language Models (LLMs) store a vast repository of semantic memories. In humans, efficient and strategic access to this memory store is a critical foundation for a variety of cognitive functions. Such access has long been a focus of psychology and the computational mechanisms behind it are now well characterized. Much of this understanding has been gleaned from a widely-used neuropsychological and cognitive science assessment called the Semantic Fluency Task (SFT), which requires the generation of as many semantically constrained concepts as possible. Our goal is to apply mechanistic interpretability techniques to bring greater rigor to the study of semantic memory foraging in LLMs. To this end, we present preliminary results examining SFT as a case study. A central focus is on convergent and divergent patterns of generative memory search, which in humans play complementary strategic roles in efficient memory foraging. We show that these same behavioral signatures, critical to human performance on the SFT, also emerge as identifiable patterns in LLMs across distinct layers. Potentially, this analysis provides new insights into how LLMs may be adapted into closer cognitive alignment with humans, or alternatively, guided toward productive cognitive \emph{disalignment} to enhance complementary strengths in human-AI interaction.

---

### Deep Learning for Financial Time Series: A Large-Scale Benchmark of Risk-Adjusted Performance

Authors: Adir Saly-Kaufmann, Kieran Wood, Jan Peter-Calliess, Stefan Zohren
Published: 2026-03-02
Categories: q-fin.TR, cs.LG
arXiv: https://arxiv.org/abs/2603.01820v1

We present a large scale benchmark of modern deep learning architectures for a financial time series prediction and position sizing task, with a primary focus on Sharpe ratio optimization. Evaluating linear models, recurrent networks, transformer based architectures, state space models, and recent sequence representation approaches, we assess out of sample performance on a daily futures dataset spanning commodities, equity indices, bonds, and FX spanning 2010 to 2025. Our evaluation goes beyond average returns and includes statistical significance, downside and tail risk measures, breakeven transaction cost analysis, robustness to random seed selection, and computational efficiency. We find that models explicitly designed to learn rich temporal representations consistently outperform linear benchmarks and generic deep learning models, which often lead the ranking in standard time series benchmarks. Hybrid models such as VSN with LSTM, a combination of Variable Selection Networks (VSN) and LSTMs, achieves the highest overall Sharpe ratio, while VSN with xLSTM and LSTM with PatchTST exhibit superior downside adjusted characteristics. xLSTM demonstrates the largest breakeven transaction cost buffer, indicating improved robustness to trading frictions.

---

### Neural Operator-Grounded Continuous Tensor Function Representation and Its Applications

Authors: Ruoyang Su, Xi-Le Zhao, Sheng Liu, Wei-Hao Wu, Yisi Luo, Michael K. Ng
Published: 2026-03-02
Categories: cs.CV, math.NA
arXiv: https://arxiv.org/abs/2603.01812v1

Recently, continuous tensor functions have attracted increasing attention, because they can unifiedly represent data both on mesh grids and beyond mesh grids. However, since mode-$n$ product is essentially discrete and linear, the potential of current continuous tensor function representations is still locked. To break this bottleneck, we suggest neural operator-grounded mode-$n$ operators as a continuous and nonlinear alternative of discrete and linear mode-$n$ product. Instead of mapping the discrete core tensor to the discrete target tensor, proposed mode-$n$ operator directly maps the continuous core tensor function to the continuous target tensor function, which provides a genuine continuous representation of real-world data and can ameliorate discretization artifacts. Empowering with continuous and nonlinear mode-$n$ operators, we propose a neural operator-grounded continuous tensor function representation (abbreviated as NO-CTR), which can more faithfully represent complex real-world data compared with classic discrete tensor representations and continuous tensor function representations. Theoretically, we also prove that any continuous tensor function can be approximated by NO-CTR. To examine the capability of NO-CTR, we suggest an NO-CTR-based multi-dimensional data completion model. Extensive experiments across various data on regular mesh grids (multi-spectral images and color videos), on mesh girds with different resolutions (Sentinel-2 images) and beyond mesh grids (point clouds) demonstrate the superiority of NO-CTR.

---

### GCTAM: Global and Contextual Truncated Affinity Combined Maximization Model For Unsupervised Graph Anomaly Detection

Authors: Xiong Zhang, Hong Peng, Zhenli He, Cheng Xie, Xin Jin, Hua Jiang
Published: 2026-03-02
Categories: cs.SI, cs.GR, cs.LG
arXiv: https://arxiv.org/abs/2603.01806v1

Anomalies often occur in real-world information networks/graphs, such as malevolent users, malicious comments, banned users, and fake news in social graphs. The latest graph anomaly detection methods use a novel mechanism called truncated affinity maximization (TAM) to detect anomaly nodes without using any label information and achieve impressive results. TAM maximizes the affinities among the normal nodes while truncating the affinities of the anomalous nodes to identify the anomalies. However, existing TAM-based methods truncate suspicious nodes according to a rigid threshold that ignores the specificity and high-order affinities of different nodes. This inevitably causes inefficient truncations from both normal and anomalous nodes, limiting the effectiveness of anomaly detection. To this end, this paper proposes a novel truncation model combining contextual and global affinity to truncate the anomalous nodes. The core idea of the work is to use contextual truncation to decrease the affinity of anomalous nodes, while global truncation increases the affinity of normal nodes. Extensive experiments on massive real-world datasets show that our method surpasses peer methods in most graph anomaly detection tasks. In highlights, compared with previous state-of-the-art methods, the proposed method has +15\% $\sim$ +20\% improvements in two famous real-world datasets, Amazon and YelpChi. Notably, our method works well in large datasets, Amazin-all and YelpChi-all, and achieves the best results, while most previous models cannot complete the tasks.

---

### Non-verbal Real-time Human-AI Interaction in Constrained Robotic Environments

Authors: Dragos Costea, Alina Marcu, Cristina Lazar, Marius Leordeanu
Published: 2026-03-02
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2603.01804v1

We study the ongoing debate regarding the statistical fidelity of AI-generated data compared to human-generated data in the context of non-verbal communication using full body motion. Concretely, we ask if contemporary generative models move beyond surface mimicry to participate in the silent, but expressive dialogue of body language. We tackle this question by introducing the first framework that generates a natural non-verbal interaction between Human and AI in real-time from 2D body keypoints. Our experiments utilize four lightweight architectures which run at up to 100 FPS on an NVIDIA Orin Nano, effectively closing the perception-action loop needed for natural Human-AI interaction. We trained on 437 human video clips and demonstrated that pretraining on synthetically-generated sequences reduces motion errors significantly, without sacrificing speed. Yet, a measurable reality gap persists. When the best model is evaluated on keypoints extracted from cutting-edge text-to-video systems, such as SORA and VEO, we observe that performance drops on SORA-generated clips. However, it degrades far less on VEO, suggesting that temporal coherence, not image fidelity, drives real-world performance. Our results demonstrate that statistically distinguishable differences persist between Human and AI motion.

---

### What Papers Don't Tell You: Recovering Tacit Knowledge for Automated Paper Reproduction

Authors: Lehui Li, Ruining Wang, Haochen Song, Yaoxin Mao, Tong Zhang, Yuyao Wang, Jiayi Fan, Yitong Zhang, Jieping Ye, Chengqi Zhang, Yongshun Gong
Published: 2026-03-02
Categories: cs.AI
arXiv: https://arxiv.org/abs/2603.01801v1

Automated paper reproduction -- generating executable code from academic papers -- is bottlenecked not by information retrieval but by the tacit knowledge that papers inevitably leave implicit. We formalize this challenge as the progressive recovery of three types of tacit knowledge -- relational, somatic, and collective -- and propose \method, a graph-based agent framework with a dedicated mechanism for each: node-level relation-aware aggregation recovers relational knowledge by analyzing implementation-unit-level reuse and adaptation relationships between the target paper and its citation neighbors; execution-feedback refinement recovers somatic knowledge through iterative debugging driven by runtime signals; and graph-level knowledge induction distills collective knowledge from clusters of papers sharing similar implementations. On an extended ReproduceBench spanning 3 domains, 10 tasks, and 40 recent papers, \method{} achieves an average performance gap of 10.04\% against official implementations, improving over the strongest baseline by 24.68\%. The code will be publicly released upon acceptance; the repository link will be provided in the final version.

---

### Phase-Type Variational Autoencoders for Heavy-Tailed Data

Authors: Abdelhakim Ziani, András Horváth, Paolo Ballarini
Published: 2026-03-02
Categories: cs.LG, cs.AI, stat.ML, stat.OT
arXiv: https://arxiv.org/abs/2603.01800v2

Heavy-tailed distributions are ubiquitous in real-world data, where rare but extreme events dominate risk and variability. However, standard Variational Autoencoders (VAEs) employ simple decoder distributions, such as Gaussian distributions, that fail to capture heavy-tailed behavior, while existing heavy-tail-aware extensions remain restricted to predefined parametric families whose tail behavior is fixed a priori. We propose the Phase-Type Variational Autoencoder (PH-VAE), whose decoder distribution is a latent-conditioned Phase-Type (PH) distribution, defined as the absorption time of a continuous-time Markov chain (CTMC). This formulation composes multiple exponential time scales, yielding a flexible and analytically tractable decoder that adapts its finite-range tail behavior directly from the observed data. Experiments on synthetic and real-world benchmarks demonstrate that PH-VAE accurately approximates diverse heavy-tailed distributions, significantly outperforming Gaussian, Student-t, and extreme-value-based VAE decoders in modeling observed tail behavior and extreme quantiles. In multivariate settings, PH-VAE captures realistic cross-dimensional tail dependence through its shared latent representation. To our knowledge, this is the first work to integrate Phase-Type distributions into deep generative modeling, bridging applied probability and representation learning.

---

### Incremental, inconsistency-resilient reasoning over Description Logic Abox streams

Authors: Cas Proost, Pieter Bonte
Published: 2026-03-02
Categories: cs.AI, cs.LO
arXiv: https://arxiv.org/abs/2603.01799v1

More and more, data is being produced in a streaming fashion. This has led to increased interest into how actionable insights can be extracted in real time from data streams through Stream Reasoning. Reasoning over data streams raises multiple challenges, notably the high velocity of data, the real time requirement of the reasoning, and the noisy and volatile nature of streams. This paper proposes novel semantics for incremental reasoning over streams of Description Logic ABoxes, in order to tackle these challenges. To address the first two challenges, our semantics for reasoning over sliding windows on streams allow for incrementally computing the materialization of the window based on the materialization of the previous window. Furthermore, to deal with the volatile nature of streams, we present novel semantics for inconsistency repair on such windows, based on preferred repair semantics. We then detail our proposed semi-naive algorithms for incremental materialization maintenance in the case of OWL2 RL, both in the presence of inconsistencies and without.

---

### PleaSQLarify: Visual Pragmatic Repair for Natural Language Database Querying

Authors: Robin Shing Moon Chan, Rita Sevastjanova, Mennatallah El-Assady
Published: 2026-03-02
Categories: cs.HC, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2603.01795v1

Natural language database interfaces broaden data access, yet they remain brittle under input ambiguity. Standard approaches often collapse uncertainty into a single query, offering little support for mismatches between user intent and system interpretation. We reframe this challenge through pragmatic inference: while users economize expressions, systems operate on priors over the action space that may not align with the users'. In this view, pragmatic repair -- incremental clarification through minimal interaction -- is a natural strategy for resolving underspecification. We present \textsc{PleaSQLarify}, which operationalizes pragmatic repair by structuring interaction around interpretable decision variables that enable efficient clarification. A visual interface complements this by surfacing the action space for exploration, requesting user disambiguation, and making belief updates traceable across turns. In a study with twelve participants, \textsc{PleaSQLarify} helped users recognize alternative interpretations and efficiently resolve ambiguity. Our findings highlight pragmatic repair as a design principle that fosters effective user control in natural language interfaces.

---

### ALTER: Asymmetric LoRA for Token-Entropy-Guided Unlearning of LLMs

Authors: Xunlei Chen, Jinyu Guo, Yuang Li, Zhaokun Wang, Yi Gong, Jie Zou, Jiwei Wei, Wenhong Tian
Published: 2026-03-02
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2603.01792v1

Large language models (LLMs) have advanced to encompass extensive knowledge across diverse domains. Yet controlling what a LLMs should not know is important for ensuring alignment and thus safe use. However, effective unlearning in LLMs is difficult due to the fuzzy boundary between knowledge retention and forgetting. This challenge is exacerbated by entangled parameter spaces from continuous multi-domain training, often resulting in collateral damage, especially under aggressive unlearning strategies. Furthermore, the computational overhead required to optimize State-of-the-Art (SOTA) models with billions of parameters poses an additional barrier. In this work, we present ALTER, a lightweight unlearning framework for LLMs to address both the challenges of knowledge entanglement and unlearning efficiency. ALTER operates through two phases: (I) high entropy tokens are captured and learned via the shared A matrix in LoRA, followed by (II) an asymmetric LoRA architecture that achieves a specified forgetting objective by parameter isolation and unlearning tokens within the target subdomains. Serving as a new research direction for achieving unlearning via token-level isolation in the asymmetric framework. ALTER achieves SOTA performance on TOFU, WMDP, and MUSE benchmarks with over 95% forget quality and shows minimal side effects through preserving foundational tokens. By decoupling unlearning from LLMs' billion-scale parameters, this framework delivers excellent efficiency while preserving over 90% of model utility, exceeding baseline preservation rates of 47.8-83.6%.

---

### Semantic Novelty Trajectories in 80,000 Books: A Cross-Corpus Embedding Analysis

Authors: Fred Zimmerman
Published: 2026-03-02
Categories: cs.CL, cs.IR
arXiv: https://arxiv.org/abs/2603.01791v1

I apply Schmidhuber's compression progress theory of interestingness at corpus scale, analyzing semantic novelty trajectories in more than 80,000 books spanning two centuries of English-language publishing. Using sentence-transformer paragraph embeddings and a running-centroid novelty measure, I compare 28,730 pre-1920 Project Gutenberg books (PG19) against 52,796 modern English books (Books3, approximately 1990-2010). The principal findings are fourfold. First, mean paragraph-level novelty is roughly 10% higher in modern books (0.503 vs. 0.459). Second, trajectory circuitousness -- the ratio of cumulative path length to net displacement in embedding space -- nearly doubles in the modern corpus (+67%). Third, convergent narrative curves, in which novelty declines toward a settled semantic register, are 2.3x more common in pre-1920 literature. Fourth, novelty is orthogonal to reader quality ratings (r = -0.002), suggesting that interestingness in Schmidhuber's sense is structurally independent of perceived literary merit. Clustering paragraph-level trajectories via PAA-16 representations reveals eight distinct narrative-shape archetypes whose distribution shifts substantially between eras. All analysis code and an interactive exploration toolkit are publicly available at https://bigfivekiller.online/novelty_hub.

---

### nchellwig at SemEval-2026 Task 3: Self-Consistent Structured Generation (SCSG) for Dimensional Aspect-Based Sentiment Analysis using Large Language Models

Authors: Nils Constantin Hellwig, Jakob Fehle, Udo Kruschwitz, Christian Wolff
Published: 2026-03-02
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.01788v1

We present Self-Consistent Structured Generation (SCSG) for Dimensional Aspect-Based Sentiment Analysis in SemEval-2026 Task 3 (Track A). SCSG enhances prediction reliability by executing a LoRA-adapted large language model multiple times per instance, retaining only tuples that achieve a majority consensus across runs. To mitigate the computational overhead of multiple forward passes, we leverage vLLM's PagedAttention mechanism for efficient key--value cache reuse. Evaluation across 6 languages and 8 language--domain combinations demonstrates that self-consistency with 15 executions yields statistically significant improvements over single-inference prompting, with our system (leveraging Gemma 3) ranking in the top seven across all settings, achieving second place on three out of four English subsets and first place on Tatar-Restaurant for DimASTE.

---

### Learning Shortest Paths with Generative Flow Networks

Authors: Nikita Morozov, Ian Maksimov, Daniil Tiapkin, Sergey Samsonov
Published: 2026-03-02
Categories: cs.LG, cs.AI, stat.ML
arXiv: https://arxiv.org/abs/2603.01786v1

In this paper, we present a novel learning framework for finding shortest paths in graphs utilizing Generative Flow Networks (GFlowNets). First, we examine theoretical properties of GFlowNets in non-acyclic environments in relation to shortest paths. We prove that, if the total flow is minimized, forward and backward policies traverse the environment graph exclusively along shortest paths between the initial and terminal states. Building on this result, we show that the pathfinding problem in an arbitrary graph can be solved by training a non-acyclic GFlowNet with flow regularization. We experimentally demonstrate the performance of our method in pathfinding in permutation environments and in solving Rubik's Cubes. For the latter problem, our approach shows competitive results with state-of-the-art machine learning approaches designed specifically for this task in terms of the solution length, while requiring smaller search budget at test-time.

---

### Co-Evolutionary Multi-Modal Alignment via Structured Adversarial Evolution

Authors: Guoxin Shi, Haoyu Wang, Zaihui Yang, Yuxing Wang, Yongzhe Chang
Published: 2026-03-02
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2603.01784v1

Adversarial behavior plays a central role in aligning large language models with human values. However, existing alignment methods largely rely on static adversarial settings, which fundamentally limit robustness, particularly in multimodal settings with a larger attack surface. In this work, we move beyond static adversarial supervision and introduce co-evolutionary alignment with evolving attacks, instantiated by CEMMA (Co-Evolutionary Multi-Modal Alignment), an automated and adaptive framework for multimodal safety alignment. We introduce an Evolutionary Attacker that decomposes adversarial prompts into method templates and harmful intents. By employing genetic operators, including mutation, crossover, and differential evolution, it enables simple seed attacks to inherit the structural efficacy of sophisticated jailbreaks. The Adaptive Defender is iteratively updated on the synthesized hard negatives, forming a closed-loop process that adapts alignment to evolving attacks. Experiments show that the Evolutionary Attacker substantially increases red-teaming jailbreak attack success rate (ASR), while the Adaptive Defender improves robustness and generalization across benchmarks with higher data efficiency, without inducing excessive benign refusal, and remains compatible with inference-time defenses such as AdaShield.

---

### GAM-RAG: Gain-Adaptive Memory for Evolving Retrieval in Retrieval-Augmented Generation

Authors: Yifan Wang, Mingxuan Jiang, Zhihao Sun, Yixin Cao, Yicun Liu, Keyang Chen, Guangnan Ye, Hongfeng Chai
Published: 2026-03-02
Categories: cs.AI
arXiv: https://arxiv.org/abs/2603.01783v1

Retrieval-Augmented Generation (RAG) grounds large language models with external evidence, but many implementations rely on pre-built indices that remain static after construction. Related queries therefore repeat similar multi-hop traversal, increasing latency and compute. Motivated by schema-based learning in cognitive neuroscience, we propose GAM-RAG, a training-free framework that accumulates retrieval experience from recurring or related queries and updates retrieval memory over time. GAM-RAG builds a lightweight, relation-free hierarchical index whose links capture potential co-occurrence rather than fixed semantic relations. During inference, successful retrieval episodes provide sentence-level feedback, updating sentence memories so evidence useful for similar reasoning types becomes easier to activate later. To balance stability and adaptability under noisy feedback, we introduce an uncertainty-aware, Kalman-inspired gain rule that jointly updates memory states and perplexity-based uncertainty estimates. It applies fast updates for reliable novel signals and conservative refinement for stable or noisy memories. We provide a theoretical analysis of the update dynamics, and empirically show that GAM-RAG improves average performance by 3.95% over the strongest baseline and by 8.19% with 5-turn memory, while reducing inference cost by 61%. Our code and datasets are available at: https://anonymous.4open.science/r/GAM_RAG-2EF6.

---

### D3LM: A Discrete DNA Diffusion Language Model for Bidirectional DNA Understanding and Generation

Authors: Zhao Yang, Hengchang Liu, Chuan Cao, Bing Su
Published: 2026-03-02
Categories: cs.LG, q-bio.GN
arXiv: https://arxiv.org/abs/2603.01780v1

Early DNA foundation models adopted BERT-style training, achieving good performance on DNA understanding tasks but lacking generative capabilities. Recent autoregressive models enable DNA generation, but employ left-to-right causal modeling that is suboptimal for DNA where regulatory relationships are inherently bidirectional. We present D3LM (\textbf{D}iscrete \textbf{D}NA \textbf{D}iffusion \textbf{L}anguage \textbf{M}odel), which unifies bidirectional representation learning and DNA generation through masked diffusion. D3LM directly adopts the Nucleotide Transformer (NT) v2 architecture but reformulates the training objective as masked diffusion in discrete DNA space, enabling both bidirectional understanding and generation capabilities within a single model. Compared to NT v2 of the same size, D3LM achieves improved performance on understanding tasks. Notably, on regulatory element generation, D3LM achieves an SFID of 10.92, closely approaching real DNA sequences (7.85) and substantially outperforming the previous best result of 29.16 from autoregressive models. Our work suggests diffusion language models as a promising paradigm for unified DNA foundation models. We further present the first systematic study of masked diffusion models in the DNA domain, investigating practical design choices such as tokenization schemes and sampling strategies, thereby providing empirical insights and a solid foundation for future research. D3LM has been released at https://huggingface.co/collections/Hengchang-Liu/d3lm.

---

### LLM-as-an-Annotator: Training Lightweight Models with LLM-Annotated Examples for Aspect Sentiment Tuple Prediction

Authors: Nils Constantin Hellwig, Jakob Fehle, Udo Kruschwitz, Christian Wolff
Published: 2026-03-02
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.01778v1

Training models for Aspect-Based Sentiment Analysis (ABSA) tasks requires manually annotated data, which is expensive and time-consuming to obtain. This paper introduces LA-ABSA, a novel approach that leverages Large Language Model (LLM)-generated annotations to fine-tune lightweight models for complex ABSA tasks. We evaluate our approach on five datasets for Target Aspect Sentiment Detection (TASD) and Aspect Sentiment Quad Prediction (ASQP). Our approach outperformed previously reported augmentation strategies and achieved competitive performance with LLM-prompting in low-resource scenarios, while providing substantial energy efficiency benefits. For example, using 50 annotated examples for in-context learning (ICL) to guide the annotation of unlabeled data, LA-ABSA achieved an F1 score of 49.85 for ASQP on the SemEval Rest16 dataset, closely matching the performance of ICL prompting with Gemma-3-27B (51.10), while requiring significantly lower computational resources.

---

### FreeAct: Freeing Activations for LLM Quantization

Authors: Xiaohao Liu, Xiaobo Xia, Manyi Zhang, Ji-Fu Li, Xianzhi Yu, Fei Shen, Xiu Su, See-Kiong Ng, Tat-Seng Chua
Published: 2026-03-02
Categories: cs.CL, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2603.01776v2

Quantization is pivotal for mitigating the significant memory and computational overhead of Large Language Models (LLMs). While emerging transformation-based methods have successfully enhanced quantization by projecting feature spaces onto smoother manifolds using orthogonal matrices, they typically enforce a rigid one-to-one transformation constraint. This static approach fails to account for the dynamic patterns inherent in input activations, particularly within diffusion LLMs (dLLMs) and Multimodal LLMs (MLLMs), where varying token types exhibit distinct distributions. To advance this, we propose FreeAct, a novel quantization framework that relaxes the static one-to-one constraint to accommodate dynamic activation disparities. Theoretically, we leverage the rank-deficient nature of activations to derive a solution space that extends beyond simple inverse matrices, enabling the decoupling of activation transformations from weights. Methodologically, FreeAct identifies token-specific dynamics (i.e., vision v.s. text, or masked tokens) and allocates distinct transformation matrices to the activation side, while maintaining a unified, static transformation for the weights. Extensive experiments across dLLMs and MLLMs demonstrate that FreeAct significantly outperforms baselines, up to 5.3% performance improvement, with in-depth analyses. Our code will be publicly released.

---

### Beyond the Resumé: A Rubric-Aware Automatic Interview System for Information Elicitation

Authors: Harry Stuart, Masahiro Kaneko, Timothy Baldwin
Published: 2026-03-02
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.01775v1

Effective hiring is integral to the success of an organisation, but it is very challenging to find the most suitable candidates because expert evaluation (e.g.\ interviews conducted by a technical manager) are expensive to deploy at scale. Therefore, automated resume scoring and other applicant-screening methods are increasingly used to coarsely filter candidates, making decisions on limited information. We propose that large language models (LLMs) can play the role of subject matter experts to cost-effectively elicit information from each candidate that is nuanced and role-specific, thereby improving the quality of early-stage hiring decisions. We present a system that leverages an LLM interviewer to update belief over an applicant's rubric-oriented latent traits in a calibrated way. We evaluate our system on simulated interviews and show that belief converges towards the simulated applicants' artificially-constructed latent ability levels. We release code, a modest dataset of public-domain/anonymised resumes, belief calibration tests, and simulated interviews, at \href{https://github.com/mbzuai-nlp/beyond-the-resume}{https://github.com/mbzuai-nlp/beyond-the-resume}. Our demo is available at \href{https://btr.hstu.net}{https://btr.hstu.net}.

---

### AnnoABSA: A Web-Based Annotation Tool for Aspect-Based Sentiment Analysis with Retrieval-Augmented Suggestions

Authors: Nils Constantin Hellwig, Jakob Fehle, Udo Kruschwitz, Christian Wolff
Published: 2026-03-02
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.01773v1

We introduce AnnoABSA, the first web-based annotation tool to support the full spectrum of Aspect-Based Sentiment Analysis (ABSA) tasks. The tool is highly customizable, enabling flexible configuration of sentiment elements and task-specific requirements. Alongside manual annotation, AnnoABSA provides optional Large Language Model (LLM)-based retrieval-augmented generation (RAG) suggestions that offer context-aware assistance in a human-in-the-loop approach, keeping the human annotator in control. To improve prediction quality over time, the system retrieves the ten most similar examples that are already annotated and adds them as few-shot examples in the prompt, ensuring that suggestions become increasingly accurate as the annotation process progresses. Released as open-source software under the MIT License, AnnoABSA is freely accessible and easily extendable for research and practical applications.

---

### Hyperparameter Trajectory Inference with Conditional Lagrangian Optimal Transport

Authors: Harry Amad, Mihaela van der Schaar
Published: 2026-03-02
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2603.01771v3

Neural networks (NNs) often have critical behavioural trade-offs that are set at design time with hyperparameters-such as reward weights in reinforcement learning or quantile targets in regression. Post-deployment, however, user preferences can evolve, making initial settings undesirable, necessitating potentially expensive retraining. To circumvent this, we introduce the task of Hyperparameter Trajectory Inference (HTI): to learn, from observed data, how a NN's conditional output distribution changes with its hyperparameters, and construct a surrogate model that approximates the NN at unobserved hyperparameter settings. HTI requires extending existing trajectory inference approaches to incorporate conditions, exacerbating the challenge of ensuring inferred paths are feasible. We propose an approach based on conditional Lagrangian optimal transport, jointly learning the Lagrangian function governing hyperparameter-induced dynamics along with the associated optimal transport maps and geodesics between observed marginals, which form the surrogate model. We incorporate inductive biases based on the manifold hypothesis and least-action principles into the learned Lagrangian, improving surrogate model feasibility. We empirically demonstrate that our approach reconstructs NN outputs across various hyperparameter spectra better than other alternatives.

---

### CHLU: The Causal Hamiltonian Learning Unit as a Symplectic Primitive for Deep Learning

Authors: Pratik Jawahar, Maurizio Pierini
Published: 2026-03-02
Categories: cs.LG, cs.AI, physics.app-ph
arXiv: https://arxiv.org/abs/2603.01768v2

Current deep learning primitives dealing with temporal dynamics suffer from a fundamental dichotomy: they are either discrete and unstable (LSTMs) \citep{pascanu_difficulty_2013}, leading to exploding or vanishing gradients; or they are continuous and dissipative (Neural ODEs) \citep{dupont_augmented_2019}, which destroy information over time to ensure stability. We propose the \textbf{Causal Hamiltonian Learning Unit} (pronounced: \textit{clue}), a novel Physics-grounded computational learning primitive. By enforcing a Relativistic Hamiltonian structure and utilizing symplectic integration, a CHLU strictly conserves phase-space volume, as an attempt to solve the memory-stability trade-off. We show that the CHLU is designed for infinite-horizon stability, as well as controllable noise filtering. We then demonstrate a CHLU's generative ability using the MNIST dataset as a proof-of-principle.

---

### Downstream Task Inspired Underwater Image Enhancement: A Perception-Aware Study from Dataset Construction to Network Design

Authors: Bosen Lin, Feng Gao, Yanwei Yu, Junyu Dong, Qian Du
Published: 2026-03-02
Categories: cs.CV, eess.IV
arXiv: https://arxiv.org/abs/2603.01767v1

In real underwater environments, downstream image recognition tasks such as semantic segmentation and object detection often face challenges posed by problems like blurring and color inconsistencies. Underwater image enhancement (UIE) has emerged as a promising preprocessing approach, aiming to improve the recognizability of targets in underwater images. However, most existing UIE methods mainly focus on enhancing images for human visual perception, frequently failing to reconstruct high-frequency details that are critical for task-specific recognition. To address this issue, we propose a Downstream Task-Inspired Underwater Image Enhancement (DTI-UIE) framework, which leverages human visual perception model to enhance images effectively for underwater vision tasks. Specifically, we design an efficient two-branch network with task-aware attention module for feature mixing. The network benefits from a multi-stage training framework and a task-driven perceptual loss. Additionally, inspired by human perception, we automatically construct a Task-Inspired UIE Dataset (TI-UIED) using various task-specific networks. Experimental results demonstrate that DTI-UIE significantly improves task performance by generating preprocessed images that are beneficial for downstream tasks such as semantic segmentation, object detection, and instance segmentation. The codes are publicly available at https://github.com/oucailab/DTIUIE.

---

### StreamTTO: Efficient Online Test-Time Optimization for Video Depth Completion

Authors: Minseok Seo, Wonjun Lee, Jaehyuk Jang, Changick Kim
Published: 2026-03-02
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.01765v5

Monocular depth foundation models generalize across diverse scenes, but recovering accurate metric depth consistent with a target sensor remains challenging under sensor variation and domain shift. Test-time optimization (TTO) guided by sparse depth addresses this limitation, yet independent per-frame adaptation incurs substantial computational cost. We propose StreamTTO, an online test-time optimization framework for video depth completion that reuses visual features and adaptation state. A frozen feature extractor with causal temporal attention incorporates past visual context into features cached for repeated decoder optimization. Causal Sliding-Window Optimization directly updates a shared depth decoder using cached features and sparse observations from recent frames, requiring only two passes over the window per incoming frame after initialization. Reusing recent observations and the adapted decoder substantially reduces optimization steps. We also introduce MS-Depth, comprising approximately 1.45M synchronized RGB--LiDAR frames from 40 hours of recordings. Its uninterrupted sequences capture transitions among indoor, underground, and outdoor environments under daytime and nighttime conditions, enabling evaluation of adaptation to changes in illumination, scene structure, and depth range. Experiments on KITTI, Bonn, NYUv2 Raw, TUM RGB-D, DDAD, and MS-Depth demonstrate competitive depth completion accuracy. On MS-Depth, StreamTTO achieves average processing speeds of approximately 15~FPS with VGGT and 25~FPS with MoGe-2 at an input resolution of 518 X 392.

---

### Efficient Test-Time Optimization for Depth Completion via Low-Rank Decoder Adaptation

Authors: Minseok Seo, Wonjun Lee, Jaehyuk Jang, Changick Kim
Published: 2026-03-02
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.01765v4

Zero-shot depth completion has gained attention for its ability to generalize across environments without sensor-specific datasets or retraining. However, most existing approaches rely on diffusion-based test-time optimization, which is computationally expensive due to iterative denoising. Recent visual-prompt-based methods reduce training cost but still require repeated forward--backward passes through the full frozen network to optimize input-level prompts, resulting in slow inference. In this work, we show that adapting only the decoder is sufficient for effective test-time optimization, as depth foundation models concentrate depth-relevant information within a low-dimensional decoder subspace. Based on this insight, we propose a lightweight test-time adaptation method that updates only this low-dimensional subspace using sparse depth supervision. Our approach achieves state-of-the-art performance, establishing a new Pareto frontier between accuracy and efficiency for test-time adaptation. Extensive experiments on five indoor and outdoor datasets demonstrate consistent improvements over prior methods, highlighting the practicality of fast zero-shot depth completion.

---

### DGNet: Discrete Green Networks for Data-Efficient Learning of Spatiotemporal PDEs

Authors: Yingjie Tan, Quanming Yao, Yaqing Wang
Published: 2026-03-02
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.01762v1

Spatiotemporal partial differential equations (PDEs) underpin a wide range of scientific and engineering applications. Neural PDE solvers offer a promising alternative to classical numerical methods. However, existing approaches typically require large numbers of training trajectories, while high-fidelity PDE data are expensive to generate. Under limited data, their performance degrades substantially, highlighting their low data efficiency. A key reason is that PDE dynamics embody strong structural inductive biases that are not explicitly encoded in neural architectures, forcing models to learn fundamental physical structure from data. A particularly salient manifestation of this inefficiency is poor generalization to unseen source terms. In this work, we revisit Green's function theory-a cornerstone of PDE theory-as a principled source of structural inductive bias for PDE learning. Based on this insight, we propose DGNet, a discrete Green network for data-efficient learning of spatiotemporal PDEs. The key idea is to transform the Green's function into a graph-based discrete formulation, and embed the superposition principle into the hybrid physics-neural architecture, which reduces the burden of learning physical priors from data, thereby improving sample efficiency. Across diverse spatiotemporal PDE scenarios, DGNet consistently achieves state-of-the-art accuracy using only tens of training trajectories. Moreover, it exhibits robust zero-shot generalization to unseen source terms, serving as a stress test that highlights its data-efficient structural design.

---

### Position: Modular Memory is the Key to Continual Learning Agents

Authors: Vaggelis Dorovatas, Malte Schwerin, Andrew D. Bagdanov, Lucas Caccia, Antonio Carta, Laurent Charlin, Barbara Hammer, Tyler L. Hayes, Timm Hess, Christopher Kanan, Dhireesha Kudithipudi, Xialei Liu et al.
Published: 2026-03-02
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2603.01761v2

Foundation models have transformed machine learning through large-scale pretraining and increased test-time compute. Despite surpassing human performance in several domains, these models remain fundamentally limited in continuous operation, experience accumulation, and personalization, capabilities that are central to adaptive intelligence. While continual learning research has long targeted these goals, its historical focus on in-weight learning (IWL), i.e., updating a single model's parameters to absorb new knowledge, has rendered catastrophic forgetting a persistent challenge. Our position is that combining the strengths of In-Weight Learning (IWL) and the newly emerged capabilities of In-Context Learning (ICL) through the design of modular memory is the missing piece for continual adaptation at scale. We outline a conceptual framework for modular memory-centric architectures that leverage ICL for rapid adaptation and knowledge accumulation, and IWL for stable updates to model capabilities, charting a practical roadmap toward continually learning agents.

---

### Meta-Learning Hyperparameters for Parameter Efficient Fine-Tuning

Authors: Zichen Tian, Yaoyao Liu, Qianru Sun
Published: 2026-03-02
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.01759v1

Training large foundation models from scratch for domain-specific applications is almost impossible due to data limits and long-tailed distributions -- taking remote sensing (RS) as an example. Fine-tuning natural image pre-trained models on RS images is a straightforward solution. To reduce computational costs and improve performance on tail classes, existing methods apply parameter-efficient fine-tuning (PEFT) techniques, such as LoRA and AdaptFormer. However, we observe that fixed hyperparameters -- such as intra-layer positions, layer depth, and scaling factors, can considerably hinder PEFT performance, as fine-tuning on RS images proves highly sensitive to these settings. To address this, we propose MetaPEFT, a method incorporating adaptive scalers that dynamically adjust module influence during fine-tuning. MetaPEFT dynamically adjusts three key factors of PEFT on RS images: module insertion, layer selection, and module-wise learning rates, which collectively control the influence of PEFT modules across the network. We conduct extensive experiments on three transfer-learning scenarios and five datasets in both RS and natural image domains. The results show that MetaPEFT achieves state-of-the-art performance in cross-spectral adaptation, requiring only a small amount of trainable parameters and improving tail-class accuracy significantly.

---

### Unifying Heterogeneous Multi-Modal Remote Sensing Detection Via Language-Pivoted Pretraining

Authors: Yuxuan Li, Yuming Chen, Yunheng Li, Ming-Ming Cheng, Xiang Li, Jian Yang
Published: 2026-03-02
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.01758v1

Heterogeneous multi-modal remote sensing object detection aims to accurately detect objects from diverse sensors (e.g., RGB, SAR, Infrared). Existing approaches largely adopt a late alignment paradigm, in which modality alignment and task-specific optimization are entangled during downstream fine-tuning. This tight coupling complicates optimization and often results in unstable training and suboptimal generalization. To address these limitations, we propose BabelRS, a unified language-pivoted pretraining framework that explicitly decouples modality alignment from downstream task learning. BabelRS comprises two key components: Concept-Shared Instruction Aligning (CSIA) and Layerwise Visual-Semantic Annealing (LVSA). CSIA aligns each sensor modality to a shared set of linguistic concepts, using language as a semantic pivot to bridge heterogeneous visual representations. To further mitigate the granularity mismatch between high-level language representations and dense detection objectives, LVSA progressively aggregates multi-scale visual features to provide fine-grained semantic guidance. Extensive experiments demonstrate that BabelRS stabilizes training and consistently outperforms state-of-the-art methods without bells and whistles. Code: https://github.com/zcablii/SM3Det.

---

### StepVAR: Structure-Texture Guided Pruning for Visual Autoregressive Models

Authors: Keli Liu, Zhendong Wang, Wengang Zhou, Houqiang Li
Published: 2026-03-02
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.01757v1

Visual AutoRegressive (VAR) models based on next-scale prediction enable efficient hierarchical generation, yet the inference cost grows quadratically at high resolutions. We observe that the computationally intensive later scales predominantly refine high-frequency textures and exhibit substantial spatial redundancy, in contrast to earlier scales that determine the global structural layout. Existing pruning methods primarily focus on high-frequency detection for token selection, often overlooking structural coherence and consequently degrading global semantics. To address this limitation, we propose StepVAR, a training-free token pruning framework that accelerates VAR inference by jointly considering structural and textural importance. Specifically, we employ a lightweight high-pass filter to capture local texture details, while leveraging Principal Component Analysis (PCA) to preserve global structural information. This dual-criterion design enables the model to retain tokens critical for both fine-grained fidelity and overall composition. To maintain valid next-scale prediction under sparse tokens, we further introduce a nearest neighbor feature propagation strategy to reconstruct dense feature maps from pruned representations. Extensive experiments on state-of-the-art text-to-image and text-to-video VAR models demonstrate that StepVAR achieves substantial inference speedups while maintaining generation quality. Quantitative and qualitative evaluations consistently show that our method outperforms existing acceleration approaches, validating its effectiveness and general applicability across diverse VAR architectures.

---

### NeuroSymb-MRG: Differentiable Abductive Reasoning with Active Uncertainty Minimization for Radiology Report Generation

Authors: Rong Fu, Yiqing Lyu, Chunlei Meng, Muge Qi, Yabin Jin, Qi Zhao, Li Bao, Juntao Gao, Fuqian Shi, Nilanjan Dey, Wei Luo, Simon Fong
Published: 2026-03-02
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.01756v2

Automatic generation of radiology reports seeks to reduce clinician workload while improving documentation consistency. Existing methods that adopt encoder-decoder or retrieval-augmented pipelines achieve progress in fluency but remain vulnerable to visual-linguistic biases, factual inconsistency, and lack of explicit multi-hop clinical reasoning. We present NeuroSymb-MRG, a unified framework that integrates NeuroSymbolic abductive reasoning with active uncertainty minimization to produce structured, clinically grounded reports. The system maps image features to probabilistic clinical concepts, composes differentiable logic-based reasoning chains, decodes those chains into templated clauses, and refines the textual output via retrieval and constrained language-model editing. An active sampling loop driven by rule-level uncertainty and diversity guides clinician-in-the-loop adjudication and promptbook refinement. Experiments on standard benchmarks demonstrate consistent improvements in factual consistency and standard language metrics compared to representative baselines.

---

### Federated Agentic AI for Wireless Networks: Fundamentals, Approaches, and Applications

Authors: Lingyi Cai, Yu Zhang, Ruichen Zhang, Yinqiu Liu, Tao Jiang, Dusit Niyato, Wei Ni, Abbas Jamalipour
Published: 2026-03-02
Categories: cs.NI, cs.AI
arXiv: https://arxiv.org/abs/2603.01755v1

Agentic artificial intelligence (AI) presents a promising pathway toward realizing autonomous and self-improving wireless network services. However, resource-constrained, widely distributed, and data-heterogeneous nature of wireless networks poses significant challenges to existing agentic AI that relies on centralized architectures, leading to high communication overhead, privacy risks, and non-independent and identically distributed (non-IID) data. Federated learning (FL) has the potential to improve the overall loop of agentic AI through collaborative local learning and parameter sharing without exchanging raw data. This paper proposes new federated agentic AI approaches for wireless networks. We first summarize fundamentals of agentic AI and mainstream FL types. Then, we illustrate how each FL type can strengthen a specific component of agentic AI's loop. Moreover, we conduct a case study on using FRL to improve the performance of agentic AI's action decision in low-altitude wireless networks (LAWNs). Finally, we provide a conclusion and discuss future research directions.

---

### Causal Circuit Tracing Reveals Distinct Computational Architectures in Single-Cell Foundation Models: Inhibitory Dominance, Biological Coherence, and Cross-Model Convergence

Authors: Ihor Kendiukhov
Published: 2026-03-02
Categories: cs.LG, q-bio.CB, q-bio.GN
arXiv: https://arxiv.org/abs/2603.01752v2

Motivation: Sparse autoencoders (SAEs) decompose foundation model activations into interpretable features, but causal feature-to-feature interactions across network depth remain unknown for biological foundation models. Results: We introduce causal circuit tracing by ablating SAE features and measuring downstream responses, and apply it to Geneformer V2-316M and scGPT whole-human across four conditions (96,892 edges, 80,191 forward passes). Both models show approximately 53 percent biological coherence and 65 to 89 percent inhibitory dominance, invariant to architecture and cell type. scGPT produces stronger effects (mean absolute d = 1.40 vs. 1.05) with more balanced dynamics. Cross-model consensus yields 1,142 conserved domain pairs (10.6x enrichment, p < 0.001). Disease-associated domains are 3.59x more likely to be consensus. Gene-level CRISPRi validation shows 56.4 percent directional accuracy, confirming co-expression rather than causal encoding.

---

### Shape-Interpretable Visual Self-Modeling Enables Geometry-Aware Continuum Robot Control

Authors: Peng Yu, Xin Wang, Ning Tan
Published: 2026-03-02
Categories: cs.RO, cs.AI, cs.LG, eess.SY
arXiv: https://arxiv.org/abs/2603.01751v1

Continuum robots possess high flexibility and redundancy, making them well suited for safe interaction in complex environments, yet their continuous deformation and nonlinear dynamics pose fundamental challenges to perception, modeling, and control. Existing vision-based control approaches often rely on end-to-end learning, achieving shape regulation without explicit awareness of robot geometry or its interaction with the environment. Here, we introduce a shape-interpretable visual self-modeling framework for continuum robots that enables geometry-aware control. Robot shapes are encoded from multi-view planar images using a Bezier-curve representation, transforming visual observations into a compact and physically meaningful shape space that uniquely characterizes the robot's three-dimensional configuration. Based on this representation, neural ordinary differential equations are employed to self-model both shape and end-effector dynamics directly from data, enabling hybrid shape-position control without analytical models or dense body markers. The explicit geometric structure of the learned shape space allows the robot to reason about its body and surroundings, supporting environment-aware behaviors such as obstacle avoidance and self-motion while maintaining end-effector objectives. Experiments on a cable-driven continuum robot demonstrate accurate shape-position regulation and tracking, with shape errors within 1.56% of image resolution and end-effector errors within 2% of robot length, as well as robust performance in constrained environments. By elevating visual shape representations from two-dimensional observations to an interpretable three-dimensional self-model, this work establishes a principled alternative to vision-based end-to-end control and advances autonomous, geometry-aware manipulation for continuum robots.

---

### Practical Deep Heteroskedastic Regression

Authors: Mikkel Jordahn, Jonas Vestergaard Jensen, James Harrison, Michael Riis Andersen, Mikkel N. Schmidt
Published: 2026-03-02
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.01750v1

Uncertainty quantification (UQ) in deep learning regression is of wide interest, as it supports critical applications including sequential decision making and risk-sensitive tasks. In heteroskedastic regression, where the uncertainty of the target depends on the input, a common approach is to train a neural network that parameterizes the mean and the variance of the predictive distribution. Still, training deep heteroskedastic regression models poses practical challenges in the trade-off between uncertainty quantification and mean prediction, such as optimization difficulties, representation collapse, and variance overfitting. In this work we identify previously undiscussed fallacies and propose a simple and efficient procedure that addresses these challenges jointly by post-hoc fitting a variance model across the intermediate layers of a pretrained network on a hold-out dataset. We demonstrate that our method achieves on-par or state-of-the-art uncertainty quantification on several molecular graph datasets, without compromising mean prediction accuracy and remaining cheap to use at prediction time.

---

### Discrete World Models via Regularization

Authors: Davide Bizzaro, Luciano Serafini
Published: 2026-03-02
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2603.01748v1

World models aim to capture the states and dynamics of an environment in a compact latent space. Moreover, using Boolean state representations is particularly useful for search heuristics and symbolic reasoning and planning. Existing approaches keep latents informative via decoder-based reconstruction, or instead via contrastive or reward signals. In this work, we introduce Discrete World Models via Regularization (DWMR): a reconstruction-free and contrastive-free method for unsupervised Boolean world-model learning. In particular, we introduce a novel world-modeling loss that couples latent prediction with specialized regularizers. Such regularizers maximize the entropy and independence of the representation bits through variance, correlation, and coskewness penalties, while simultaneously enforcing a locality prior for sparse action changes. To enable effective optimization, we also introduce a novel training scheme improving robustness to discrete roll-outs. Experiments on two benchmarks with underlying combinatorial structure show that DWMR learns more accurate representations and transitions than reconstruction-based alternatives. Finally, DWMR can also be paired with an auxiliary reconstruction decoder, and this combination yields additional gains.

---

### An Analysis of Multi-Task Architectures for the Hierarchic Multi-Label Problem of Vehicle Model and Make Classification

Authors: Alexandru Manole, Laura Diosan
Published: 2026-03-02
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2603.01746v1

Most information in our world is organized hierarchically; however, many Deep Learning approaches do not leverage this semantically rich structure. Research suggests that human learning benefits from exploiting the hierarchical structure of information, and intelligent models could similarly take advantage of this through multi-task learning. In this work, we analyze the advantages and limitations of multi-task learning in a hierarchical multi-label classification problem: car make and model classification. Considering both parallel and cascaded multi-task architectures, we evaluate their impact on different Deep Learning classifiers (CNNs, Transformers) while varying key factors such as dropout rate and loss weighting to gain deeper insight into the effectiveness of this approach. The tests are conducted on two established benchmarks: StanfordCars and CompCars. We observe the effectiveness of the multi-task paradigm on both datasets, improving the performance of the investigated CNN in almost all scenarios. Furthermore, the approach yields significant improvements on the CompCars dataset for both types of models.

---

### Action-Guided Attention for Video Action Anticipation

Authors: Tsung-Ming Tai, Sofia Casarin, Andrea Pilzer, Werner Nutt, Oswald Lanz
Published: 2026-03-02
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.01743v2

Anticipating future actions in videos is challenging, as the observed frames provide only evidence of past activities, requiring the inference of latent intentions to predict upcoming actions. Existing transformer-based approaches, which rely on dot-product attention over pixel representations, often lack the high-level semantics necessary to model video sequences for effective action anticipation. As a result, these methods tend to overfit to explicit visual cues present in the past frames, limiting their ability to capture underlying intentions and degrading generalization to unseen samples. To address this, we propose Action-Guided Attention (AGA), an attention mechanism that explicitly leverages predicted action sequences as queries and keys to guide sequence modeling. Our approach fosters the attention module to emphasize relevant moments from the past based on the upcoming activity and combine this information with the current frame embedding via a dedicated gating function. The design of AGA enables post-training analysis of the knowledge discovered from the training set. Experiments on the widely adopted EPIC-Kitchens-100 benchmark demonstrate that AGA generalizes well from validation to unseen test sets. Post-training analysis can further examine the action dependencies captured by the model and the counterfactual evidence it has internalized, offering transparent and interpretable insights into its anticipative predictions.

---

### Rethinking Policy Diversity in Ensemble Policy Gradient in Large-Scale Reinforcement Learning

Authors: Naoki Shitanda, Motoki Omura, Tatsuya Harada, Takayuki Osa
Published: 2026-03-02
Categories: cs.LG, cs.AI, cs.RO
arXiv: https://arxiv.org/abs/2603.01741v2

Scaling reinforcement learning to tens of thousands of parallel environments requires overcoming the limited exploration capacity of a single policy. Ensemble-based policy gradient methods, which employ multiple policies to collect diverse samples, have recently been proposed to promote exploration. However, merely broadening the exploration space does not always enhance learning capability, since excessive exploration can reduce exploration quality or compromise training stability. In this work, we theoretically analyze the impact of inter-policy diversity on learning efficiency in policy ensembles, and propose Coupled Policy Optimization which regulates diversity through KL constraints between policies. The proposed method enables effective exploration and outperforms strong baselines such as SAPG, PBT, and PPO across multiple tasks, including challenging dexterous manipulation, in terms of both sample efficiency and final performance. Furthermore, analysis of policy diversity and effective sample size during training reveals that follower policies naturally distribute around the leader, demonstrating the emergence of structured and efficient exploratory behavior. Our results indicate that diverse exploration under appropriate regulation is key to achieving stable and sample-efficient learning in ensemble policy gradient methods. Project page at https://naoki04.github.io/paper-cpo/ .

---

### CA-AFP: Cluster-Aware Adaptive Federated Pruning

Authors: Om Govind Jha, Harsh Shukla, Haroon R. Lone
Published: 2026-03-02
Categories: cs.LG, cs.AI, cs.DC
arXiv: https://arxiv.org/abs/2603.01739v1

Federated Learning (FL) faces major challenges in real-world deployments due to statistical heterogeneity across clients and system heterogeneity arising from resource-constrained devices. While clustering-based approaches mitigate statistical heterogeneity and pruning techniques improve memory and communication efficiency, these strategies are typically studied in isolation. We propose CA-AFP, a unified framework that jointly addresses both challenges by performing cluster-specific model pruning. In CA-AFP, clients are first grouped into clusters, and a separate model for each cluster is adaptively pruned during training. The framework introduces two key innovations: (1) a cluster-aware importance scoring mechanism that combines weight magnitude, intra-cluster coherence, and gradient consistency to identify parameters for pruning, and (2) an iterative pruning schedule that progressively removes parameters while enabling model self-healing through weight regrowth. We evaluate CA-AFP on two widely used human activity recognition benchmarks, UCI HAR and WISDM, under natural user-based federated partitions. Experimental results demonstrate that CA-AFP achieves a favorable balance between predictive accuracy, inter-client fairness, and communication efficiency. Compared to pruning-based baselines, CA-AFP consistently improves accuracy and lower performance disparity across clients with limited fine-tuning, while requiring substantially less communication than dense clustering-based methods. It also shows robustness to different Non-IID levels of data. Finally, ablation studies analyze the impact of clustering, pruning schedules and scoring mechanism offering practical insights into the design of efficient and adaptive FL systems.

---

### Bootstrapping Embeddings for Low Resource Languages

Authors: Merve Basoz, Andrew Horne, Mattia Opper
Published: 2026-03-02
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.01732v2

Embedding models are crucial to modern NLP. However, the creation of the most effective models relies on carefully constructed supervised finetuning data. For high resource languages, such as English, such datasets are readily available. However, for hundreds of other languages, they are simply non-existent. We investigate whether the advent of large language models can help to bridge this gap. We test three different strategies for generating synthetic triplet data used to optimise embedding models. These include in-context learning as well as two novel approaches, leveraging adapter composition and cross lingual finetuning of the LLM generator (XL-LoRA) respectively. We find that while in-context learning still falls short of strong non-synthetic baselines, adapter composition and XL-LoRA yield strong performance gains across a wide array of tasks and languages, offering a clear, scalable pathway to producing performant embedding models for a wide variety of languages.

---