Knowledge
-
Abductive World Modeling via Causal Representation Learning
2026-09-29 08:18:27Ziqi Liu, Songhan Yang, Linfan Zhou, Jiatong Liu, Lijun Peng, Long Wan, Yinqi BaiAbstract
The central challenge of world modeling is to learn representations that capture how the world evolves. However, existing world models predominantly represent future states without explicitly capturing the latent causes underlying their evolution, limiting their ability to reason about why and how the world changes. To address this limitation, we propose Abductive World Modeling (AWM), a framework that learns structured causal representations by abductively inferring latent causes from predicted futures. Specifically, we realize AWM through the Hierarchical Abductive State Pyramid (HASP), which organizes the inferred world state into three complementary components - Entity, Dynamic, and Relation - capturing what exists, how it changes, and how entities interact, respectively. By jointly reasoning over the current observation and its predicted future, HASP abductively infers these latent factors and integrates them into a structured state representation for downstream reasoning. To the best of our knowledge, AWM is the first framework to introduce abductive state inference into latent-space world modeling for learning structured representations of world dynamics. Experiments across physical prediction, causal reasoning, and action understanding demonstrate the effectiveness of our approach. Compared with V-JEPA, a state-of-the-art latent-space world model, AWM improves physical prediction AUROC by 10.7%, causal reasoning accuracy by 16.8%, and action Top-1 accuracy by 68.0%.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36985
PDF
-
SRJudge: Empowering Large Language Models with Selective Reasoning for Fine-Grained Knowledge Concept Tagging
2026-09-29 08:16:20Zhiwei Yang, Jiahua Yang, Huiru Lin, Xing Chen, Quanlong GuanAbstract
Knowledge concept tagging aims to assign specific concept or topic labels to educational content, which is essential for both educators and learners in traditional and online teaching practices. Recent work has explored large language models (LLMs) for this task, achieving promising performance. However, LLMs still struggle to select the correct concept from a large-scale candidate set due to the high dimensionality of the decision space. In this paper, we propose a novel three-stage Select-Reason-Judge (SRJudge) framework, which empowers LLMs with selective reasoning capability for fine-grained knowledge concept tagging. Specifically, the Selector in Stage 1 first narrows the candidate concepts to a top-K shortlist by fine-tuning a small language model (SLM), e.g., BERT, since the top-$K$ predictions hit the correct concept in most cases, thereby reducing the decision space of correct candidates. Next, the Stage 2 Reasoner employs a lightweight LLM for refined reasoning over the shortlisted candidates. It further integrates an improved reinforcement learning strategy with a dynamic task-specific reward function and a pruning mechanism to better align with human reasoning preferences. Finally, a larger LLM acts as a judger that evaluates the overall rationality of the reasoning process and its explanations to determine the final output. In addition, we construct two high-quality datasets for further validation, i.e., the biology dataset S_Bio and the physics dataset S_Phy. Experimental results demonstrate that our method consistently outperforms state-of-the-art baselines across benchmark datasets, verifying its effectiveness and superiority. Resources are available at: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36982
PDF
-
Beyond Readability: Evaluating Task Information Recoverability
Abstract
Direct visual readability and task-information recoverability are different quantities. Failure to decode a target from a fixed observation need not eliminate access to that target through another recovery route. We develop an evaluation perspective that makes the observation, query, target, and available knowledge explicit and measures the overlap between routes' success sets. For information available on the original visible surface under suitable imaging conditions, direct optical recovery reads the target from the image, optionally after restoration; entity-linked recovery uses residual visual evidence to identify the depicted entity and accesses its target through an entity--attribute relation in a specified knowledge resource. Such access can draw on stored knowledge or an external source. A controlled book-cover study instantiates external access with a fixed title--author catalog, comparing optical author recovery with visual title resolution and deterministic lookup under resolution degradation. Entity-linked successes persist across the tested vision--language models, revealing information access beyond the tested direct visual frontier despite substantial differences in absolute performance. A substantial optical-only region remains. These complementary outcomes show why visual degradation should be evaluated through the task information accessible along specified routes and knowledge resources, alongside direct readability.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36957
PDF
-
ER-JEPA: Experience Replay Improves Joint-Embedding Predictive Learning in Language Models
2026-09-29 07:56:01Jingnan Pu, Zi-En Fan, Feng LianAbstract
Large language models (LLMs) excel at token-level generation but may learn undesirable abstract semantics and lack comprehensive perception. LLM-JEPA mitigates this by aligning different views of the same underlying knowledge via a joint-embedding predictive architecture (JEPA). However, strong alignment does not necessarily lead to accurate, stable predictions. To address this, we propose ER-JEPA, which adds an episodic replay path to LLM-JEPA. ER-JEPA stores training pairs in a memory. At each step, it stores and retrieves relevant data to provide additional supervision. This enables learning from both the current batch and stored training pairs, providing additional supervision for token prediction and representation alignment. Experiments across multiple datasets (NL-RX, GSM8K, Spider, and NQ-Open) demonstrate that ER-JEPA consistently outperforms LLM-JEPA.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36952
PDF
-
Forensic-Aware Continual Adaptation for Image Forgery Localization
2026-09-29 07:47:57Chenqi Kong, Song Xia, Anwei Luo, Peisong He, Alex C. Kot, Yuming FangAbstract
The rapid evolution of image manipulation techniques has raised growing public security concerns. Existing Image Forgery Localization (IFL) methods can accurately localize manipulated regions but are often unable to adapt to newly emerging forgeries. In real-world forensic scenarios, data typically arrive sequentially, yet continual model adaptation remains largely unexplored in IFL. To bridge this gap, we introduce the first continual learning framework for IFL and establish a comprehensive benchmark under two realistic data-evolution protocols: cross-dataset and cross-content continual learning. Evaluations of representative state-of-the-art IFL and continual learning methods reveal substantial performance degradation, highlighting two key challenges: (1) adaptively capturing intrinsic forensic traces from incoming data across unseen domains, and (2) preserving previously acquired forensic knowledge during sequential adaptation. To address these challenges, we propose a forensic-aware continual adaptation framework. First, a forensic trace mining module employs Spatial Mixture-of-Forensic-Experts (SMoFE) to dynamically route complementary forensic cues across spatial locations, together with Forensic Evidence-Guided Dense Prompting (FEGDP) to transform low-level forensic traces into structured localization evidence for SAM. Second, Fisher-weighted LoRA Gradient (FLAG) surgery identifies old-task-sensitive adaptation directions and suppresses conflicting updates, mitigating catastrophic forgetting while preserving plasticity for emerging forgery domains. Extensive experiments demonstrate state-of-the-art performance in both pixel-level forgery localization and image-level forgery detection across diverse continual learning scenarios.
Abstract (translated)
URL
https://arxiv.org/abs/2609.38251
PDF
-
Track-and-Complete: Learning Humanoid Skills from a Single Failed Human Video
Abstract
Learning humanoid skills from videos typically requires a successful human demonstration, which often demands custom data collection. Although failures have traditionally been treated only as negative examples in robot learning, they can still reveal a usable trajectory prefix before the task fails, as well as the intended outcome. To leverage this information from a failed-attempt video, we propose TRACC, a pipeline that imitates the useful portion of the motion trajectory and then completes the task based on the inferred task outcome. The usable motion prefix serves as prior knowledge until the failure occurs, after which the task-completion reward guides the policy to learn the intended task goal without requiring a successful task trajectory. We evaluate our method on six in-the-wild failed human tasks from the Oops! dataset. Our experimental results demonstrate the effectiveness of the proposed approach for learning from failed attempts when no successful demonstration is available. Thus, these findings establish failed human videos as a viable source of supervision for humanoid skill learning.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36924
PDF
-
Do Gaussian Scenes Contain Enough Structure for Intrinsic Segmentation?
2026-09-29 07:26:59Mohamed Rayan Barhdadi, Hasan Yazar, Erchin Serpedin, Mehmet Tuncel, Hasan KurbanAbstract
Gaussian segmentation is usually posed as transferring object knowledge from 2D foundation models into a 3D representation. This leaves a fundamental question unanswered: how much object structure is already encoded by a trained gaussian scene? We investigate this question with GS-IntSeg, an intrinsic, mask-free, and training-free method that constructs partitions using only: gaussian geometry, opacity, spherical-harmonic radiance, and deformation trajectories. On the dynamic Neu3D and HyperNeRF datasets, GS-IntSeg obtains a mean of 0.677 mIoU across multi-view and monocular scenes without masks, external features, or segmentation training. This intrinsic formulation also enables GS-IntSeg to require approximately 2.5 minutes per HyperNeRF scene on a consumer RTX 5080 to construct its partitions, over 10x faster than SAM-based methods that require mask generation, feature rendering, and other stages. These results suggest that gaussians alone can approach mask-supervised performance in gaussian scenes segmentation. While a gap remains between intrinsic and foundation model based methods, to our knowledge, GS-IntSeg is the first mask-free approach to gaussian segmentation, motivating a re-evaluation of the assumption that gaussian scene segmentation must be based on external masks and pointing toward an alternative faster, more generalizable segmentation approach.
Abstract (translated)
URL
https://arxiv.org/abs/2606.18623
PDF
-
RareDx: Controlled Knowledge Integration and Graph-Grounded Policy Optimization for Rare-Disease Diagnosis
2026-09-29 07:14:46Bo Zhang, Yuchen Wang, Dongbai Li, Matthew Yu Heng Wong, Qingkai Zeng, Lijun Wang, Tien-Yin Wong, Peng Cui, Tianyu LiuAbstract
Rare-disease diagnosis is a long-tail reasoning problem: phenotypes are incomplete, individual disorders are sparsely documented, and relevant evidence is distributed across ontologies, gene annotations, and biomedical text. Language models consequently favor common conditions, miss rare candidates, or produce plausible but invalid names. We introduce RareDx, which couples controlled evidence use with knowledge-graph-grounded policy optimization. RareDx-Harness normalizes heterogeneous records into one ranked-diagnosis task and compares direct inference, static retrieval, adaptive tools, and structured phenotype-gene-disease reasoning over a shared knowledge layer. The training pipeline combines Top-10 post-training with RareDx-KGPO, our knowledge-graph-grounded policy optimization method. Its reward projects predictions into a canonical disease graph and integrates curated graded relevance, ontology proximity, biomedical similarity, and phenotype consistency. Vocabulary and output-budget constraints prevent dense partial credit from rewarding fabricated or overlong differentials. Across eight benchmarks, the complete RareDx system centered on Qwen3.5-9B reaches 38.34 macro Hit@10, 1.60 points above GPT-5.5 under the archived protocol; a disjoint validation-selection audit retains a 6.80-point routing gain over Direct on held-out cases. The 27B system reaches 23.53/36.56/40.76 at Hit@1/5/10. Controlled ablations show that retrieval is not uniformly helpful and that controlled routing is central to the gain. These results indicate that structured medical knowledge can turn a compact model into a competitive diagnostic ranker across heterogeneous long-tail settings in clinical practice.
Abstract (translated)
URL
https://arxiv.org/abs/2609.35549
PDF
-
State Trace Rationale As Auxiliary Task in Reinforcement Learning
2026-09-29 07:07:22Muhammad U. Nasir, Alex Vogt, Steven D. James, Julian TogeliusAbstract
We propose STRAT, an auxiliary task that trains deep reinforcement learning (RL) agents to predict a short textual trace of their own state. Inspired by human spatial navigation, the description combines landmark, route, and survey knowledge, tracking the agent's position, inventory, goals, and immediate progress. Environment rules generate this text online without human labelling. Our method adds a single auxiliary head to a standard policy. Across 60 sparse-reward XLand-MiniGrid tasks, STRAT solves complex environments where standard RL fails outright, while compacting state representations and preventing rank collapse. Beyond performance gains, the predicted trace provides a readable account of agent beliefs at every step for no extra cost.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36867
PDF
-
You Cannot Recover What Was Never Measured: Quantifying the Information Ceiling of Ultra-Low-Field MRI Super-Resolution
2026-09-29 06:46:38Prathamesh Pradeep Khole, Shreya Handa, Utkarsh Gupta, Razvan MarinescuAbstract
Generative super-resolution models can turn portable 64 mT MRI into images that look like 3T scans, and the field evaluates them with PSNR, SSIM, and pixelwise uncertainty, most often on pairs built by synthetically degrading high-field images. Prior work acknowledges that these models hallucinate and that the problem is ill posed, but to our knowledge no study measures how much information about the individual subject the real low-field scan actually contains. We measure it. Using paired 64 mT and 3T scans of the same subjects from three public datasets, and a measurement protocol validated on tests whose correct answer is known in advance, we find that, judged over the whole brain, real 64 mT scans carry structure specific to the individual only down to approximately 3 to 4 mm half-pitch in plane, and coarser still through plane. Standard synthetic degradations preserve subject information roughly 1 mm beyond this ceiling, so models trained and benchmarked on synthetic pairs are evaluated on information that real scanners never record. We then test trained diffusion models and a publicly released external model on real paired acquisitions; 24 trained runs of five architectures (GAN, diffusion, and transformer families) give the coverage of the audit. On every subject where faithfulness can be measured, fine output detail is no more correlated with the subject's own 3T scan than with a stranger's, while sample-variance uncertainty does not distinguish fabricated structure from reconstruction difficulty. Because PSNR and SSIM score resemblance to a reference rather than whether detail belongs to the subject, a benchmark scored by them cannot tell recovery from fabrication. Code for the measurement protocol will be released so that recoverability claims can be tested for newer models.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36837
PDF
-
RED: Reconstruction Evolution Dynamics for Generalizable AI-Generated Image Detection
2026-09-29 06:33:54Wenpeng Mu, Junshan Jin, Tanfeng Sun, Xinghao Jiang, Qiang XuAbstract
The rapid evolution of image generators calls for forensic cues that generalize beyond known generation mechanisms. Existing detectors often rely on static image representations or endpoint reconstruction discrepancies, leaving the evolution of intermediate reconstruction stages underexplored. We observe that the relative token predictability of real and generated images can reverse across reconstruction scales, suggesting that intermediate stages may expose forensic evidence overlooked by endpoint comparisons. Motivated by this observation, we propose RED (Reconstruction Evolution Dynamics), a framework that captures transferable forensic cues from coarse-to-fine reconstruction evolution. To our knowledge, RED is the first framework to use scale-wise token predictability to guide forensic evidence aggregation across intermediate reconstruction states. It represents the reconstruction trajectory produced by a frozen multiscale VQ-VAE in the shared feature space of a frozen CLIP encoder. To connect the observed predictability variations with visual evidence, RED learns image-adaptive stage weights from scale-wise token negative log-likelihoods provided by a frozen VAR model. A cross-stage evidence aggregation module then jointly models the original-image representation and the weighted reconstruction features, capturing complementary forensic cues through interactions along the reconstruction trajectory. Experiments on six diverse benchmarks demonstrate that RED achieves the highest average accuracy of 92.5\% and average precision of 97.5\% among the evaluated methods. Further evaluations show strong robustness to common image degradations, supporting the value of reconstruction evolution for generalizable AI-generated image detection. The code will be made publicly available upon acceptance of this paper.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36822
PDF
-
Beyond Conditional Independence: Root Cause Analysis with Deep Causal Models
2026-09-29 05:43:23Md Musfiqur Rahman, Kenneth Lee, Ziwei Jiang, Padmaja Jonnalagedda, Ruocheng Guo, Murat KocaogluAbstract
Root cause analysis (RCA) is a critical problem in many real-world scenarios. RCA enables the identification of faulty or failing mechanisms in a system by comparing anomalous observations with corresponding reference (i.e., regular) observations. However, existing approaches rely either on heuristic methods or on conditional independence tests with a strong unconfoundedness assumption, and thus fail to exploit other complicated distributional constraints in the presence of latent variables. To relax these assumptions, we model the underlying system as a causal model and the anomalous system as a change in the structural functions of the same causal model. Specifically, to handle unobserved confounders, we establish an implicit connection between distributional constraint testing and root cause analysis. To adapt our approach to data generated from arbitrary causal models, we employ the deep causal model (DCM) framework, in which we design the causal model using neural networks. Finally, we illustrate how our method, RCA-DCM, can utilize different levels of partial graphical knowledge to perform RCA. We evaluate RCA-DCM against state-of-the-art baselines on simulated datasets, a physics-based causal chamber and two micro-service applications. RCA-DCM improves top-1 accuracy over the strongest baseline on both Sock Shop (0.880 vs. 0.752) and Online Boutique (0.776 vs. 0.712), and when the true root cause in the causal chamber is unobserved and acts as a latent confounder, it recovers the exact root-cause set more often than any competing method (perfect recovery rate (PRR) 0.846 vs. 0.731).
Abstract (translated)
URL
https://arxiv.org/abs/2609.36771
PDF
-
QuantMLA: Function-Aligned Dual-Path Quantization for Low-Bit MLA KV Caching
2026-09-29 05:31:42Zunhai Su, Yuxuan Sun, Jianchao Tan, Tao Zhang, Ruihan Hu, Yuchen Xie, Xunliang Cai, Ngai WongAbstract
Multi-Head Latent Attention (MLA) enables expressive multi-head attention with compact caches for its content and decoupled RoPE paths, yet cache memory still scales linearly with context length and batch size. In this work, we establish a systematic model of MLA's dual-path quantization errors, characterizing their distinct effects on attention-output distortion and explaining the pronounced amplification of RoPE-path errors. Guided by this analysis, we introduce QuantMLA, a function-aligned framework for low-bit dual-path quantization. We derive path-specific transformation spaces that preserve full-precision computation while remaining fully fusible into model parameters offline, eliminating online transformation overhead. Within these spaces, QuantMLA learns path-specific transformations with function-aligned objectives: attention-output reconstruction captures the content path's coupled matching and aggregation errors, while positional QK reconstruction preserves the RoPE-induced component of the attention logits and admits a theoretical bound on output distortion. Across four MLA model families, QuantMLA enables, to our knowledge, the first reported joint INT4 caching of the content and RoPE caches with minimal accuracy degradation. Further compressing the content cache to INT2 while retaining the RoPE key cache at INT4 maintains competitive performance on challenging reasoning and code benchmarks. We develop a native low-bit MLA attention kernel that integrates unpacking and dequantization directly into attention computation. The physical cache layout provides 3.59x compression at 128K context, while a cache-pressure serving workload achieves 5.168x higher whole-job output throughput than BF16. The code will be released upon acceptance.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36760
PDF
-
Dual-Mode Low-Rank Learner with Bridge-Prototype Ensemble for Vision-Language Class-Incremental Learning
2026-09-29 05:31:41Chiyuan He, Zihuan Qiu, Fanman Meng, Chao Wang, Liangjiang Chen, Linfeng Xu, Qingbo Wu, Hongliang LiAbstract
Benefiting from transferable visual-textual alignment, CLIP has been widely adopted for class-incremental learning (CIL). However, existing learners either repeatedly update components shared across tasks, leading to knowledge overwriting, or overly isolate new-task updates, hindering the reuse of CLIP's transferable knowledge and limiting plasticity. Moreover, the text-based or bimodal classifier designs still fail to effectively integrate complementary information from the visual and textual modalities. To address these challenges, we introduce DuLBE, which couples dual-mode low-rank learning with a bridge-prototype ensemble classifier for exemplar-free CIL. DuLBE allocates two visual low-rank update modes according to the gradient demand and uses gradient routing to coordinate them: a compact and rewritable shared mode is selected from historically occupied visual directions to reuse transferable knowledge, while residual modes provide low-interference channels for task-specific variations. Building on the resulting stable inter-modal structure, we further construct geodesic bridges between visual prototypes and text embeddings on the unit hypersphere, and ensemble reliable bridge prototypes to compensate for the modality-gap limitations of textual decision boundaries. Extensive experiments under multiple settings show that DuLBE achieves state-of-the-art CIL performance while retaining the high parameter efficiency of low-rank tuning.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36759
PDF
-
Generalizable Lifelong Model Editing via Preference Optimization
2026-09-29 05:19:52Dahyun Jung, Suhyune Son, Heuiseok LimAbstract
Knowledge editing enables rapid updates of specific factual knowledge in large language models (LLMs) without full retraining. However, more realistic scenarios call for a lifelong framework that handles continual updates rather than one-off modifications. In such settings, existing editing methods often overfit to target prompts, significantly degrading both the generalization of the edited knowledge and the model's general capabilities. To address this issue, we propose GLIME (Generalizable Lifelong Model Editing), which combines knowledge editing with preference optimization over generation behavior. GLIME further incorporates replay-based editing and a gradient constraint to preserve previously edited knowledge. Experimental results show that GLIME significantly improves knowledge generalization in lifelong editing settings while maintaining both editing performance and general capabilities.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36748
PDF
-
EASE: Behavior-Adaptive Skill Curation for Self-Evolving Agents
2026-09-29 05:18:24Zhen Xiong, Qiaoyu TanAbstract
Agent skills provide a lightweight mechanism for self-evolving agents to accumulate reusable procedural knowledge without updating model parameters. However, existing learned skill curators typically optimize curation without explicitly modeling downstream executor behavior. We show that this can cause systematic cross-executor degradation: curators trained with different executors perform best when paired with their own training executor, indicating that effective skill curation is executor-dependent. We formulate behavior-adaptive skill curation and introduce EASE, a framework that learns a single curator that adapts its decisions to different executor behaviors. EASE maintains an online behavioral profile of recent execution patterns and conditions the curator on this profile, the current trajectory, and retrieved skills to add, modify, or remove skills from an evolving repository. We train the shared curator jointly across multiple frozen executors with reinforcement learning, using retrieval-aware and behavior-aware temporal attribution to focus optimization on curation actions with observable downstream influence. Across ALFWorld, ScienceWorld, and WebShop, with executors ranging from Qwen3-8B/32B and GPT-OSS-120B to unseen Kimi K2.6, DeepSeek V4 Flash, and Gemini 3.5 Flash, EASE outperforms strong skill- and memory-based baselines without per-executor finetuning. EASE also maintains 34.5--41.0% fewer skills, improves skill retrieval by 36.3--38.7% and measured edit utility by 51.8--60.0%, and reduces deployment-time inference tokens by 9.1--14.5%. These results establish behavior-adaptive skill curation as an effective principle for building self-evolving agents.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36746
PDF
-
Distilling What Matters: Confidence-Aware Selective Distillation for Large Language Models
2026-09-29 05:07:47Ayan Sengupta, Vaibhav Seth, Tanmoy ChakrabortyAbstract
Knowledge Distillation (KD) trains a smaller-capacity student model to imitate a larger-capacity teacher model by matching output distributions, implicitly assuming the teacher to be a reliable oracle. In large language models (LLMs), this assumption often fails: teacher predictions can exhibit high entropy and hallucinations, causing standard KD to degrade well-calibrated student priors. We propose CaRE-KD, a confidence-gated distillation framework that replaces static objectives with uncertainty-adaptive optimization. CaRE-KD has two components: a token-level loss (CaRE-Divergence) that adaptively switches between Forward and Reverse KL divergence based on teacher--student confidence, and a batch-level epistemic rejection mechanism (Revival) that suppresses updates when the teacher is more uncertain than the student. We provide a gradient-level analysis showing how this dual-granularity design induces a conditional calibration mechanism that prior static divergences cannot reproduce. Empirically, across eight teacher--student pairs and eleven benchmarks spanning instruction following, chat alignment, code generation, and mathematical reasoning, CaRE-KD delivers consistent gains over strong baselines (Skewed-KL, $\alpha$--$\beta$ divergence). Highlights include up to $+3.2$ average ROUGE-L on instruction-following tasks, $+2.1$ pass@1 on MBPP, $+1.7$ accuracy on GSM8k, and $+1.8$ accuracy on CollegeMath over the strongest baseline, with consistent gains in LLM-as-a-judge factuality (up to $+2.5$ per task over Skewed-RKL). Revival further acts as a principled, loss-agnostic plug-in that systematically strengthens existing distillation objectives by filtering epistemically unreliable teacher supervision.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36734
PDF
-
LAURA: Knowledge Distillation for Interpretable Ambiguous Clause Identification in Legal Contracts
2026-09-29 04:45:47Amrita Singh, Aditya Joshi, Jiaojiao Jiang, Hye-young PaikAbstract
Legal contracts contain ambiguities that expose enterprises to financial and legal risks. Some ambiguities allow flexible interpretation without triggering disputes, while others lead to significant legal conflicts. This makes identification alone insufficient, and interpretable rationale analysis essential. We propose LAURA, a post-training framework for interpretable ambiguous clause identification. LAURA leverages knowledge distillation with an IRAC-Unlearning prompting technique to transfer knowledge from a teacher LLM to an open-weight student model (<=1B parameters), which is then trained using a joint objective combining classification and rationale generation losses. The framework supports both legal and non-legal stakeholders in making informed decisions about which ambiguities require further attention. Extensive experiments across 7 baselines and 7 open-weight models demonstrate that LAURA with Flan-T5 (250M) delivers state-of-the-art interpretability over all interpretable baselines while matching the identification performance of the best-performing opaque baseline.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36707
PDF
-
Video2Skill: From Streaming Experience to Reusable Embodied Skills
2026-09-29 04:32:13Jianshu Zhang, Ce Zhang, Xiyuan Yang, Chenwei Xu, Haoran Lu, Yijiang Li, Yaqi Xie, Katia P. Sycara, Han LiuAbstract
Manipulation behaviors vary widely across objects and scenes, but they share a small set of reusable skills, and planning with these skills helps embodied agents generalize to new tasks. Yet an agent can only plan with skills it knows. Recovering skills from observed experience, the inverse of planning, builds this knowledge over time and yields skill data for training future agents. Vision-Language Models (VLMs) describe individual manipulation events well, but can they organize a stream of events into reusable skills? We formulate this problem as Streaming Embodied Skill Discovery (SESD): a model watches videos in sequence and maintains a persistent skill library that shapes its later decisions. To systematically measure this ability, we introduce Video2Skill, a benchmark that covers robot tabletop manipulation and human kitchen activity and tests three core capabilities: (i) locating manipulation events in time, (ii) grouping events of the same transformation, and (iii) deciding when to reuse an existing skill or create a new one. Across 19 open-source VLMs, many models group events at near-chance level, and scale does not consistently help. Their errors depend on how perception and library updates are coupled: joint models merge distinct transformations into one skill, while models that update the library from text descriptions duplicate recurring ones. Supervised fine-tuning, including our counterfactual library-state rebalancing (CLaRe), improves grouping but exposes a deeper bottleneck: trained models consolidate familiar skills yet rarely expand the library. Their libraries stall below half the reference size, and transformations unseen in training are located in time but almost never given a new skill. Recognizing when existing skills are insufficient thus emerges as the central challenge.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36691
PDF
-
What Makes Recurrence Effective in Looped Language Models?
2026-09-29 03:43:51Xinlin Zhuang, Siyuan Wang, Imran Razzak, Weiyang LiuAbstract
Looped language models (LoopLMs) increase computational depth through parameter sharing, offering a path to scale inference computation without adding parameters. However, it remains unclear when additional recurrence is beneficial and how architectural choices affect its effectiveness. Through controlled experiments, we systematically examine (1) when recurrence helps, (2) where it should be applied, and (3) how its conditioning affects performance. Our evaluation covers inference budgets below, within, and beyond the training horizon under knowledge and reasoning tasks. (1) We find that recurrence can improve reasoning beyond the training horizon while degrading knowledge performance, but harder reasoning instances do not consistently benefit more. (2) Performance also depends on how distinct layers and recurrent iterations are allocated, showing that effective depth alone is insufficient to predict behavior. Non-recurrent output layers improve robustness to under-unrolling, while the preferred placement of input and output layers varies with inference budget. (3) Finally, we find that conventional initial-state injection offers limited robustness to varying recurrence depth. We therefore propose history-state injection as an alternative, and show that channel-wise history-state injection combined with timestep conditioning offers a low-cost and more effective design, better preserving knowledge under extended unrolling while improving robustness across inference budgets. Overall, our results clarify when recurrent computation helps, where it fails, and offer practical guidelines for designing LoopLMs across variable inference budgets.
Abstract (translated)
URL
https://arxiv.org/abs/2609.36636
PDF
