The Missing Piece in Pre-trained Model Evaluation: Reward-Guided Decoding Unlocks Task-Oriented Behavior Without Parameter Updates

With the rapid progress of large language models (LLMs), reliably evaluating the capabilities of pre-trained LLMs has become increasingly important. The challenge is that base pre-trained models are optimized for next-token prediction and often fail to follow instructions or produce well-formed answers under standard prompting and direct decoding. As a result, benchmark performance can conflate model capability with decoding-induced failures to produce task-oriented outputs, while exposing such behavior often relies on costly post-training. Recent decodingonly approaches attempt to reshape output distributions, but such methods can be inefficient and brittle across open-ended tasks. To address these limitations, we propose Energy-Based Decoding (EBD), a training-free, reward-guided framework for activating task-oriented behaviors from frozen pre-trained LLMs across both open-ended and objective tasks. EBD augments decoding with an external lightweight reward model, steering generations toward high-utility responses while anchoring them to the pre-trained model prior through a reward-tilted target distribution. We show that EBD shifts base-model outputs toward more instructionfollowing behavior, increasing behavioral similarity to post-trained counterparts and enabling a fairer inference-time evaluation of accessible pre-trained-model behavior. Empirically, EBD outperforms baselines across five models and six benchmarks, improving Qwen3-8B-Base on AlpacaEval2.0 from 8.8 to 44.5, reducing Mistral-7B Math500 latency by 18.9x relative to prior decoding work, and remaining robust to reward-model size.

EvoVid: Temporal-Centric Self-Evolution for Video Large Language Models

Recent Video Large Language Models (Video-LLMs) have demonstrated strong capabilities in video reasoning through reinforcement learning (RL). However, existing RL pipelines rely heavily on human-annotated tasks and solutions, making them costly to scale and fundamentally constrained by human expertise. Self-evolving frameworks have recently emerged as a promising alternative through autonomous Questioner-Solver self-play. Unfortunately, these approaches are primarily designed for static modalities such as text and images, fundamentally failing to capture the temporal dynamics that are central to video reasoning. In this work, we propose $\textbf{EvoVid}$, a temporal-centric self-evolving framework that enables Video-LLMs to improve directly from raw, unannotated videos. Specifically, we introduce two complementary temporal-centric rewards: a temporal-aware Questioner reward that encourages temporally dependent question generation through temporal perturbation sensitivity, and a temporal-grounded Solver reward that provides automatic temporal supervision via inherent video segment localization. Extensive experiments across four base models and six benchmarks demonstrate consistent improvements over both base models and existing self-evolving baselines, achieving competitive performance with supervised methods. These results highlight temporal-centric self-evolution as an effective and scalable paradigm for video understanding and reasoning.

UniSurgSAM: A Unified Promptable Model for Reliable Surgical Video Segmentation

Surgical video segmentation is fundamental to computer-assisted surgery. In practice, surgeons need to dynamically specify targets throughout extended procedures, using heterogeneous cues such as visual selections, textual expressions, or audio instructions. However, existing Promptable Video Object Segmentation (PVOS) methods are typically restricted to a single prompt modality and rely on coupled frameworks that cause optimization interference between target initialization and tracking. Moreover, these methods produce hallucinated predictions when the target is absent and suffer from accumulated mask drift without failure recovery. To address these challenges, we present UniSurgSAM, a unified PVOS model enabling reliable surgical video segmentation through visual, textual, or audio prompts. Specifically, UniSurgSAM employs a decoupled two-stage framework that independently optimizes initialization and tracking to resolve the optimization interference. Within this framework, we introduce three key designs for reliability: presence-aware decoding that models target absence to suppress hallucinations; boundary-aware long-term tracking that prevents mask drift over extended sequences; and adaptive state transition that closes the loop between stages for failure recovery. Furthermore, we establish a multi-modal and multi-granular benchmark from four public surgical datasets with precise instance-level masklets. Extensive experiments demonstrate that UniSurgSAM achieves state-of-the-art performance in real time across all prompt modalities and granularities, providing a practical foundation for computer-assisted surgery. Code and datasets will be available at https://jinlab-imvr.github.io/UniSurgSAM.

Evaluating Time Awareness and Cross-modal Active Perception of Large Models via 4D Escape Room Task

Multimodal Large Language Models (MLLMs) have recently made rapid progress toward unified Omni models that integrate vision, language, and audio. However, existing environments largely focus on 2D or 3D visual context and vision-language tasks, offering limited support for temporally dependent auditory signals and selective cross-modal integration, where different modalities may provide complementary or interfering information, which are essential capabilities for realistic multimodal reasoning. As a result, whether models can actively coordinate modalities and reason under time-varying, irreversible conditions remains underexplored. To this end, we introduce \textbf{EscapeCraft-4D}, a customizable 4D environment for assessing selective cross-modal perception and time awareness in Omni models. It incorporates trigger-based auditory sources, temporally transient evidence, and location-dependent cues, requiring agents to perform spatio-temporal reasoning and proactive multimodal integration under time constraints. Building on this environment, we curate a benchmark to evaluate corresponding abilities across powerful models. Evaluation results suggest that models struggle with modality bias, and reveal significant gaps in current model's ability to integrate multiple modalities under time constraints. Further in-depth analysis uncovers how multiple modalities interact and jointly influence model decisions in complex multimodal reasoning environments.

SurgFed: Language-guided Multi-Task Federated Learning for Surgical Video Understanding

Surgical scene Multi-Task Federated Learning (MTFL) is essential for robot-assisted minimally invasive surgery (RAS) but remains underexplored in surgical video understanding due to two key challenges: (1) Tissue Diversity: Local models struggle to adapt to site-specific tissue features, limiting their effectiveness in heterogeneous clinical environments and leading to poor local predictions. (2) Task Diversity: Server-side aggregation, relying solely on gradient-based clustering, often produces suboptimal or incorrect parameter updates due to inter-site task heterogeneity, resulting in inaccurate localization. In light of these two issues, we propose SurgFed, a multi-task federated learning framework, enabling federated learning for surgical scene segmentation and depth estimation across diverse surgical types. SurgFed is powered by two appealing designs, i.e., Language-guided Channel Selection (LCS) and Language-guided Hyper Aggregation (LHA), to address the challenge of fully exploration on corss-site and cross-task. Technically, the LCS is first designed a lightweight personalized channel selection network that enhances site-specific adaptation using pre-defined text inputs, which optimally the local model learn the specific embeddings. We further introduce the LHA that employs a layer-wise cross-attention mechanism with pre-defined text inputs to model task interactions across sites and guide a hypernetwork for personalized parameter updates. Extensive empirical evidence shows that SurgFed yields improvements over the state-of-the-art methods in five public datasets across four surgical types. The code is available at https://anonymous.4open.science/r/SurgFed-070E/.

Beyond Words: Evaluating and Bridging Epistemic Divergence in User-Agent Interaction via Theory of Mind

Large Language Models (LLMs) have developed rapidly and are widely applied to both general-purpose and professional tasks to assist human users. However, they still struggle to comprehend and respond to the true user needs when intentions and instructions are imprecisely conveyed, leading to a divergence between subjective user believes and true environment states. Resolving this epistemic divergence requires Theory of Mind (ToM), yet existing ToM evaluations for LLMs primarily focus on isolated belief inference, overlooking its functional utility in real-world interaction. To this end, we formalize ToM for LLMs as a mechanism for epistemic divergence detection and resolution, and propose a benchmark, \benchname, to assess how models reconcile user beliefs and profiles in practice. Results across 11 leading models reveal a significant limitation to identify underlying cognitive gaps that impede task success. To bridge this gap, we further curate a trajectory-based ToM dataset linking belief tracking with task-related state inference. The model trained on this data via reinforcement learning shows consistent improvement in reasoning about user mental states, leading to enhanced downstream performance. Our work highlights the practical value of ToM as an essential interaction-level mechanism rather than as a standalone reasoning skill.

TimeChat-Captioner: Scripting Multi-Scene Videos with Time-Aware and Structural Audio-Visual Captions

This paper proposes Omni Dense Captioning, a novel task designed to generate continuous, fine-grained, and structured audio-visual narratives with explicit timestamps. To ensure dense semantic coverage, we introduce a six-dimensional structural schema to create "script-like" captions, enabling readers to vividly imagine the video content scene by scene, akin to a cinematographic screenplay. To facilitate research, we construct OmniDCBench, a high-quality, human-annotated benchmark, and propose SodaM, a unified metric that evaluates time-aware detailed descriptions while mitigating scene boundary ambiguity. Furthermore, we construct a training dataset, TimeChatCap-42K, and present TimeChat-Captioner-7B, a strong baseline trained via SFT and GRPO with task-specific rewards. Extensive experiments demonstrate that TimeChat-Captioner-7B achieves state-of-the-art performance, surpassing Gemini-2.5-Pro, while its generated dense descriptions significantly boost downstream capabilities in audio-visual reasoning (DailyOmni and WorldSense) and temporal grounding (Charades-STA). All datasets, models, and code will be made publicly available at https://github.com/yaolinli/TimeChat-Captioner.

TxRay: Agentic Postmortem of Live Blockchain Attacks

Decentralized Finance (DeFi) has turned blockchains into financial infrastructure, allowing anyone to trade, lend, and build protocols without intermediaries, but this openness exposes pools of value controlled by code. Within five years, the DeFi ecosystem has lost over 15.75B USD to reported exploits. Many exploits arise from permissionless opportunities that any participant can trigger using only public state and standard interfaces, which we call Anyone-Can-Take (ACT) opportunities. Despite on-chain transparency, postmortem analysis remains slow and manual: investigations start from limited evidence, sometimes only a single transaction hash, and must reconstruct the exploit lifecycle by recovering related transactions, contract code, and state dependencies. We present TxRay, a Large Language Model (LLM) agentic postmortem system that uses tool calls to reconstruct live ACT attacks from limited evidence. Starting from one or more seed transactions, TxRay recovers the exploit lifecycle, derives an evidence-backed root cause, and generates a runnable, self-contained Proof of Concept (PoC) that deterministically reproduces the incident. TxRay self-checks postmortems by encoding incident-specific semantic oracles as executable assertions. To evaluate PoC correctness and quality, we develop PoCEvaluator, an independent agentic execution-and-review evaluator. On 114 incidents from DeFiHackLabs, TxRay produces an expert-aligned root cause and an executable PoC for 105 incidents, achieving 92.11% end-to-end reproduction. Under PoCEvaluator, 98.1% of TxRay PoCs avoid hard-coding attacker addresses, a +24.8pp lift over DeFiHackLabs. In a live deployment, TxRay delivers validated root causes in 40 minutes and PoCs in 59 minutes at median latency. TxRay's oracle-validated PoCs enable attack imitation, improving coverage by 15.6% and 65.5% over STING and APE.

PathReasoner-R1: Instilling Structured Reasoning into Pathology Vision-Language Model via Knowledge-Guided Policy Optimization

Vision-Language Models (VLMs) are advancing computational pathology with superior visual understanding capabilities. However, current systems often reduce diagnosis to directly output conclusions without verifiable evidence-linked reasoning, which severely limits clinical trust and hinders expert error rectification. To address these barriers, we construct PathReasoner, the first large-scale dataset of whole-slide image (WSI) reasoning. Unlike previous work reliant on unverified distillation, we develop a rigorous knowledge-guided generation pipeline. By leveraging medical knowledge graphs, we explicitly align structured pathological findings and clinical reasoning with diagnoses, generating over 20K high-quality instructional samples. Based on the database, we propose PathReasoner-R1, which synergizes trajectory-masked supervised fine-tuning with reasoning-oriented reinforcement learning to instill structured chain-of-thought capabilities. To ensure medical rigor, we engineer a knowledge-aware multi-granular reward function incorporating an Entity Reward mechanism strictly aligned with knowledge graphs. This effectively guides the model to optimize for logical consistency rather than mere outcome matching, thereby enhancing robustness. Extensive experiments demonstrate that PathReasoner-R1 achieves state-of-the-art performance on both PathReasoner and public benchmarks across various image scales, equipping pathology models with transparent, clinically grounded reasoning capabilities. Dataset and code are available at https://github.com/cyclexfy/PathReasoner-R1.

Video-KTR: Reinforcing Video Reasoning via Key Token Attribution

Reinforcement learning (RL) has shown strong potential for enhancing reasoning in multimodal large language models, yet existing video reasoning methods often rely on coarse sequence-level rewards or single-factor token selection, neglecting fine-grained links among visual inputs, temporal dynamics, and linguistic outputs, limiting both accuracy and interpretability. We propose Video-KTR, a modality-aware policy shaping framework that performs selective, token-level RL by combining three attribution signals: (1) visual-aware tokens identified via counterfactual masking to reveal perceptual dependence; (2) temporal-aware tokens detected through frame shuffling to expose temporal sensitivity; and (3) high-entropy tokens signaling predictive uncertainty. By reinforcing only these key tokens, Video-KTR focuses learning on semantically informative, modality-sensitive content while filtering out low-value tokens. Across five challenging benchmarks, Video-KTR achieves state-of-the-art or highly competitive results, achieving 42.7\% on Video-Holmes (surpassing GPT-4o) with consistent gains on both reasoning and general video understanding tasks. Ablation studies verify the complementary roles of the attribution signals and the robustness of targeted token-level updates. Overall, Video-KTR improves accuracy and interpretability, offering a simple, drop-in extension to RL for complex video reasoning. Our code and models are available at https://github.com/zywang0104/Video-KTR.