Searcharxiv⌕ Search

arXiv subjects

Jingwei Song

Publications and source records attributed to Jingwei Song.

At least 19 recordsLinked to original sources
arXiv2026

CogniLife: Continuous Cognitive Distillation for Deployed Multi-Expert Agent Marketplaces

Multi-expert agent marketplaces -- where creators publish domain-specific AI experts for end users to consult -- are an emerging deployment pattern for language technology. A published expert can receive many requests yet little evidence about its creator, who remains active elsewhere: the cold-expert problem. We frame continuous cognitive distillation as converting attributed creator observations into revisable decision rules, communication patterns, and domain procedures. CogniLife proposes a cross-expert attribution design for this problem, with Creator Profiles for transferable creator patterns and Expert Journals for listing-specific knowledge; this design is not yet implemented or evaluated. A deployed platform partially realizes the surrounding lifecycle through creator-authorized authoring, capability-owned proposal filtering, and reversible consolidation of eligible owner-scoped memory. On ProfyBench-v1, an 85-task offline benchmark built from replay seeds and controlled internal tests, the Profy Full configuration as a whole achieves a mean overall score of 4.65/5 from the primary LLM judge, compared with 4.01 for Raw LLM and 4.28 for Expert Prompt, with the largest gains in completeness, specificity, and missing-information tasks. These configuration-level scores do not test cold-expert mitigation, cross-expert transfer, or longitudinal distillation.

cs.AI↗
arXiv2026

MemPrism: Task-Conditioned Relational Memory Views for Long-Horizon Agents

Long-horizon agents performing multi-step Web tasks must reuse earlier page observations and interactions to support subsequent actions. However, preserving relevant history does not ensure that it exposes the relations needed for the current decision. We call this gap representation mismatch. To this end, we propose MemPrism, a task-conditioned relational memory framework that separates persistent experience storage from decision-time working memory. MemPrism records interactions as the event stream and dynamically constructs relational views according to the current task context. A lightweight view policy selects the relation structure, evidence range, outcome condition, and granularity, while a deterministic composer and renderer transform historical facts into a temporary optical working-memory view for a frozen task policy. Experiments on long-horizon embodied and web-agent benchmarks show that MemPrism consistently improves the task performance, especially as trajectories become longer, while reducing memory token consumption. Furthermore, the learned view policy transfers across different VLMs without additional adaptation, demonstrating the effectiveness of task-conditioned relational views as a general memory interface for agents.

cs.AI↗
arXiv2026

From Visual Affinity to Learning Credit in Multimodal Reinforcement Learning

Reinforcement learning with verifiable rewards (RLVR) improves multimodal reasoning, but uniformly assigning a single response-level advantage to all tokens overlooks the distinct roles of visual observations and subsequent reasoning. Visual affinity alone is insufficient: predictive states may align with both image and prompt-context representations, and upweighting valid visual observations can amplify their negative-advantage updates when later reasoning fails. To address this issue, we propose \textbf{Outcome-Conditioned Visual Credit (OVC)}, a plug-in framework that separates visual-association measurement from token-level credit assignment and is adaptable to policy-optimization methods with uniform response-level advantage allocation. Relative Visual Encoding constructs a context-relative visual coordinate by contrasting each predictive state's affinity to image and prompt-context representations. Outcome-conditioned Evidence Calibration combines this coordinate with uncertainty-scaled predictive features and calibrates their contributions to outcome-discriminative evidence through mean contrasts between positive- and negative-advantage responses, scaled by regularized within-side variation. Visual association thus informs credit through observed outcome differences rather than directly determining update priority. Normalized Credit Allocation conditions the calibrated evidence on the response's advantage sign to produce positive token weights with unit global mean. These detached weights preserve advantage signs while modulating individual token-gradient contributions. OVC reuses existing rollouts without auxiliary reward models, process annotations, or inference-time modules. Instantiated through GSPO-token, OVC retains sequence-level clipping and consistently improves six-benchmark mean accuracy over GSPO across three models, with gains of 1.86-2.39 percentage points.

cs.AI↗
arXiv2026

TIDE: Trajectory-Informed Skill Distillation for Efficient Text-to-Video Prompt Optimization

Text-to-video (T2V) models are highly sensitive to prompt specificity, while iterative prompt optimization typically improves generation quality only through repeated video generation and vision-language model evaluation at inference time. We propose TIDE, a framework that distills this costly search process into an interpretable and reusable prompt-rewriting Skill. TIDE formulates prompt rewriting as a set of source-conditioned rule modules, each targeting a distinct semantic dimension. Once distilled, the frozen Skill activates the relevant rules and generates both positive and negative prompts in a single LLM call, followed by only one T2V generation, without test-time visual search or backbone fine-tuning. Evolved using only 100 prompts over 14 hours, TIDE achieves prompt-optimization performance comparable to computationally expensive search-based methods. Moreover, when transferred to previously unseen video-generation models, TIDE outperforms existing approaches specifically designed or trained for those models. Code is available at https://github.com/xiningin/TIDE_T2V.

cs.CV↗
arXiv2026

MC-EvoBench: Auditing Experience-Driven Self-Improvement in Minecraft Agents

A central challenge in embodied recursive self-improvement is enabling agents to convert interaction history into capabilities that persist beyond one world and transfer to later tasks. A saved memory or successful final action does not by itself establish that a reusable capability was acquired; evaluation must separate capability change from carried-over resources, unequal source exposure, and early execution failures. We construct MC-EvoBench as a controlled Minecraft benchmark with a fixed catalog of 19 registered evolution chains, including 12 primary evaluation chains and seven calibration or probe chains, together with three graph-aligned paired probe families for transfer. The benchmark represents each task as a task-state graph, defines stage-level environment verifiers, assigns matched source and target roles, resets the target world, and permits only a declared agent artifact to cross the source-target boundary. For each probe, no-prior, related, unrelated, and shuffled source conditions are evaluated under matched environment-step budgets and five seeds; we report TSR, WGP, and VMP and retain provenance records. The benchmark reveals a consistent direct-capability gap between controller and planner-backed variants, while related experience yields no stable transfer: it does not improve either tool probe, its small wood-species gain hinges on one seed, and all intervals include zero. Stage audits further show that early wood-collection and crafting bottlenecks often leave later capabilities unobserved. MC-EvoBench thus offers a controlled testbed for durable, transferable embodied capabilities. Code and data are available at https://github.com/xzw-ustc/mc-bench-open-source

cs.RO↗
arXiv2026

MemoWM: How World Models Change What Agents Need to Remember

Long-term agents face growing storage demands as they accumulate experience. World models capture reusable regularities that can reduce the information stored for each experience. We formulate the problem of memory allocation conditioned on a world model and introduce MemoWM, a framework that uses shared predictions to compress retained information and reconstruct omitted content. Its task-aware allocation rule balances the expected impact of reconstruction errors against storage cost, retaining information with downstream value beyond the predictive prior. Across five long-term agent-memory benchmarks, MemoWM achieves 42.42\% average answer accuracy, exceeding the strongest baseline by 2.62 percentage points, while reducing average experience-specific storage by 53.9\% relative to MIRIX, the most storage-efficient baseline. Further analysis shows that stronger world models reduce per-experience storage at comparable task quality. Accounting for model parameters reveals a trade-off between shared model capacity and recurring storage costs, with the capacity that minimizes total storage increasing as more interactions are retained. Our code is available at https://github.com/Feld-maxiu/MemoWM.

cs.LG↗
arXiv2026

ECHO-2: A Large-Scale Distributed Rollout Framework for Cost-Efficient Reinforcement Learning

Reinforcement learning (RL) is a critical stage in post-training large language models (LLMs), involving repeated interaction between rollout generation, reward evaluation, and centralized learning. Distributing rollout execution offers opportunities to leverage more cost-efficient inference resources, but introduces challenges in wide-area coordination and policy dissemination. We present ECHO-2, a distributed RL framework for post-training with remote inference workers and non-negligible dissemination latency. ECHO-2 combines centralized learning with distributed rollouts and treats bounded policy staleness as a user-controlled parameter, enabling rollout generation, dissemination, and training to overlap. We introduce an overlap-based capacity model that relates training time, dissemination latency, and rollout throughput, yielding a practical provisioning rule for sustaining learner utilization. To mitigate dissemination bottlenecks and lower cost, ECHO-2 employs peer-assisted pipelined broadcast and cost-aware activation of heterogeneous workers. Experiments on GRPO post-training of LLMs ranging from 4B to 32B parameters under real wide-area bandwidth regimes show that ECHO-2 significantly improves cost efficiency while preserving RL reward comparable to strong baselines.

cs.LG↗
arXiv2026

Beyond Token Alignment: Event Completion for Cross-Tokenizer On-Policy Distillation

On-policy distillation (OPD) transfers knowledge between language models through teacher supervision on student-generated trajectories. With different tokenizers, a single teacher token may require multiple student tokens to generate, creating intermediate states where the event is entered but not yet completed. Existing cross-tokenizer methods align tokens or text spans to construct comparable prediction targets. We study a complementary problem after partial generation: once the student produces a prefix of a teacher token, multiple next tokens may complete the same remaining bytes, but the teacher only specifies the required completion rather than how probability should be divided among these valid continuations. We introduce Event-Set Completion Distillation (ESCD), which complements cross-tokenizer probability alignment with completion-set supervision. ESCD aggregates prefix-related teacher events and supervises the total probability of byte-compatible one-step student completions, avoiding tokenizer-dependent probability splits among individual tokens. The method reuses student trajectories and predictions, requiring neither additional rollouts nor changes to the student vocabulary. Experiments demonstrate consistent gains in mathematics, code, and scientific reasoning across model families and tokenizers, extending to large-scale MoE distillation from a 1T teacher to a 35B student. Local analyses show that retaining completion sets better matches the reference supervision, while one-step completion covers over 99% of observed compatible teacher mass after partial event entry in the studied tokenizer pairs. These findings support event entry and event completion as complementary supervision targets for cross-tokenizer knowledge transfer. Code will be released on GitHub.

cs.CL↗
arXiv2026

Scaling Laws for Collapse in Asynchronous GRPO

Asynchronous reinforcement learning improves the throughput of large language model post-training by decoupling rollout generation from policy optimization, but introduces a mismatch between the behavior and learner policies. How the resulting policy staleness couples with the learning rate to govern training stability and collapse time remains poorly understood. We investigate this coupling in vanilla GRPO through controlled sweeps of the synchronization interval $S$ and constant learning rate $η$ on Llama-3.2-1B/3B, complemented by experiments on Qwen3-8B. We identify two empirical scaling laws: (i) Stability-boundary scaling: the largest stable learning rate scales approximately as $S^{-1}$, yielding a stability boundary characterized by an approximately constant product $Sη$. (ii) Collapse-time scaling: among collapsing runs, estimated collapse times scale approximately as $η^{-1}$, corresponding to a model- and setup-dependent cumulative learning-rate budget that aligns across synchronization intervals in the Llama sweeps. We interpret these laws through a local analysis of the behavior-dependent GRPO surrogate and a complementary mean-field model. Under local regularity conditions, the analysis yields an $O(Sη)$ upper bound on the staleness-induced update bias that resets at synchronization. The mean-field model shows how sufficiently strong positive feedback can sustain directional drift when update directions persist across synchronization cycles. When drift speed saturates under optimizer normalization, this mechanism predicts exit from a local surrogate-validity region after an approximately fixed cumulative learning rate. Together, these findings motivate a practical calibration rule: estimate the stability threshold and collapse budget from a coarse sweep, then jointly select $S$ and $η$ for the intended training horizon.

cs.LG↗
arXiv2026

DynaWeightPnP: Toward global real-time 3D-2D solver in PnP without correspondences

This paper addresses a special Perspective-n-Point (PnP) problem: estimating the optimal pose to align 3D and 2D shapes in real-time without correspondences, termed as correspondence-free PnP. While several studies have focused on 3D and 2D shape registration, achieving both real-time and accurate performance remains challenging. This study specifically targets the 3D-2D geometric shape registration tasks, applying the recently developed Reproducing Kernel Hilbert Space (RKHS) to address the "big-to-small" issue. An iterative reweighted least squares method is employed to solve the RKHS-based formulation efficiently. Moreover, our work identifies a unique and interesting observability issue in correspondence-free PnP: the numerical ambiguity between rotation and translation. To address this, we proposed DynaWeightPnP, introducing a dynamic weighting sub-problem and an alternative searching algorithm designed to enhance pose estimation and alignment accuracy. Experiments were conducted on a typical case, that is, a 3D-2D vascular centerline registration task within Endovascular Image-Guided Interventions (EIGIs). Results demonstrated that the proposed algorithm achieves registration processing rates of 60 Hz (without post-refinement) and 31 Hz (with post-refinement) on modern single-core CPUs, with competitive accuracy comparable to existing methods. These results underscore the suitability of DynaWeightPnP for future robot navigation tasks like EIGIs.

cs.CV↗
arXiv2026

LapaTrack-3D: 6 DoF pre-operative shape tracking for laparoscopic surgery

This work proposes a real-time 6 Degree-of-Freedom (6 DoF) tracking algorithm for monocular laparoscopic surgery. It provides alignment between intra-operative video and pre-operative data (e.g., CT). The 6 DoF tracking offers a solution for accurately locating the internal anatomy of the target organ despite the lack of tactile feedback and transparency. The ORB-SLAM2 framework is adopted and modified for prior-based 3D tracking with four major modifications. First, the primitive 3D shape is used for fast initialization of the ORB-SLAM2 monocular mode. Second, a pseudo-segmentation strategy is employed to separate the target organ from the background for tracking. Third, the 3D shape is incorporated as a geometric prior in its pose graph optimization. Fourth, the Multi-Scale Retinex with Chromaticity Preservation (MSRCP) algorithm is leveraged and modified for image enhancement in challenging illumination scenarios. In-vivo and ex-vivo experiments validate that LapaTrack-3D provides robust 3D tracking and effectively handles typical challenges such as poor illumination, fast motion, out-of-field-of-view scenarios, partial visibility, and ``organ-background'' relative motion. LapaTrack-3D achieves a processing rate of 13 Hz for 1280*720 pixel video.

cs.RO↗
arXiv2026

ToolVision: Learning When and How to Use Visual Tools with Capability-Aligned Supervision

Thinking with images allows a multimodal model to compensate for limited perception by invoking visual tools through code. Yet the prevailing SFT-then-RL recipe creates a different supervision misalignment at each stage. SFT is expected to teach how to use tools, but trajectories from stronger teachers may succeed through perceptual capabilities that a smaller student cannot reliably reproduce or exploit, causing the student to imitate tool-call patterns without learning how to make them useful. RL is expected to teach when to use tools, but outcome-only rewards make fallible tool execution a liability and suppress tool use, whereas a blanket bonus for every correct tool-using trajectory encourages valid but ineffective operations. To address these two misalignments, we introduce ToolVision. During SFT, a multi-agent pipeline explores candidate trajectories, and a committee including student-scale models scores stepwise evidence gain to rank and prune the search branches. Only successfully executed trajectories with correct final answers are retained for SFT. Before RL, ToolVision compares the learner's performance with and without tools, then rewards successful tool use only on questions where tools provide a clear benefit. Both signals are constructed automatically from public task data without additional human annotations of tool use or necessity. ToolVision-8B improves over its base on all seven main benchmarks, surpasses Thyme-7B, CodeVision-8B, and CodeDance-7B on all three high-resolution benchmarks, and outperforms Qwen3-VL-32B-Thinking on V* and HRBench 8K. We will publicly release the datasets and source code.

cs.CV↗
arXiv2026

PersonaTree: Structured Lifecycle Memory for Person Understanding in LLM Agents

Persistent LLM agents require memory representations that make the formation of person understanding explicit across long term interaction. Existing agent memory methods emphasize information retention and retrieval, yet give limited account of how accumulated interaction evidence is abstracted into person understanding. We view this process as schema formation, where situated evidence is abstracted into reusable patterns and stable person level claims. We introduce PersonaTree, a structured lifecycle memory framework that realizes this view as a three level persona tree with explicit support paths from evidence to claims. PersonaTree maintains the tree through conservative writing, confidence guided consolidation, and query conditioned path retrieval, returning only the evidence depth required by each query. Across six person understanding and persistent memory benchmarks with three answer backbones, PersonaTree ranks first in 12 of 18 compact scores and reaches the top two in 16 settings. Ablations show that hierarchy improves abstract person understanding on KnowMe, while support path retrieval improves RealPref alignment under a comparable context budget.

cs.CL↗
arXiv2026

PolarMem: A Training-Free Polarized Latent Graph Memory for Verifiable Vision-Language Models

Memory is not merely a storage mechanism for intelligent systems, but a structure for organizing evidence and constraining belief. This is especially important for multimodal reasoning, where retrieved evidence must be both query-relevant and visually consistent. However, current memory systems for vision-language models (VLMs) remain largely positive-associative: they retrieve what is similar or previously observed, but lack an explicit way to remember what has been verified as absent or logically excluded. To this end, we propose \textbf{PolarMem}, a training-free polarized latent graph memory framework for verifiable vision-language reasoning. PolarMem transforms frozen VLM perceptual signals into \textit{HAS}, \textit{NOT\_HAS}, and \textit{Uncertain} memory states through semantic consistency verification and adaptive distributional partitioning, and stores them in a polarized graph with distinct positive and negative memory relations. During inference, a lexicographical logic-aware retrieval protocol enforces logical consistency before semantic similarity, suppressing conflicting memories before they enter the model context. Across eight frozen VLM backbones and six multimodal benchmarks, PolarMem consistently improves retrieval-intensive tasks and reduces retrieval-level contradictions. These results highlight negative memory as a key mechanism for building more reliable multimodal memory systems. Our code is available at https://github.com/czs-ict/PolarMem.

cs.AI↗
arXiv2026

MineEvolve: Self-Evolution with Accumulated Knowledge for Long-Horizon Embodied Minecraft Agents

Long-horizon embodied intelligence requires agents to improve through interaction, not merely to execute plans generated from static goals. A central challenge is therefore to transform past executions into knowledge that can shape future decisions. Minecraft provides a representative testbed for this problem, where tasks such as crafting tools, building redstone components, and obtaining diamond equipment involve long prerequisite chains and are frequently disrupted by missing tools, blocked paths, GUI failures, or stagnant execution. To this end, we propose \textbf{MineEvolve}, a knowledge-driven self-evolution framework that converts execution feedback into actionable behavioral knowledge. MineEvolve first uses \underline{\emph{\textbf{\ding{182}Monitor}}} to convert each subgoal execution into typed feedback, including state changes, inventory changes, failure types, progress signals, and stagnation indicators. \underline{\emph{\textbf{\ding{183}Inducer}}} then derives reusable skills from successful executions and remedies from failed or stagnant executions. \underline{\emph{\textbf{\ding{184}Curator}}} validates, merges, filters, and retrieves these knowledge entries, while \underline{\emph{\textbf{\ding{185}Adaptor}}} uses them to repair the unfinished part of the plan under repeated failures or stagnation. Experiments on the Minecraft MCU long-horizon task suite show that MineEvolve consistently improves performance across multiple language-model planners, with larger gains on high-dependency task groups. Ablation and knowledge-accumulation studies further demonstrate that converting execution signals into structured behavioral knowledge is an effective path toward self-evolving embodied agents in long-horizon environments. Our code is available at https://github.com/xzw-ustc/MC-MineEvolve.

cs.AI↗
arXiv2026

MARS: Unleashing the Power of Speculative Decoding via Margin-Aware Verification

Speculative Decoding (SD) accelerates autoregressive large language model (LLM) inference by decoupling generation and verification. While recent methods improve draft quality by tightly coupling the drafter with the target model, the verification mechanism itself remains largely unchanged, relying on strict token-level rejection sampling. In practice, modern LLMs frequently operate in low-margin regimes where the target model exhibits weak preference among top candidates. In such cases, rejecting plausible runner-up tokens yields negligible information gain while incurring substantial rollback cost, leading to a fundamental inefficiency in verification. We propose Margin-Aware Speculative Verification, a training-free and domain-agnostic verification strategy that adapts to the target model's local decisiveness. Our method conditions verification on decision stability measured directly from the target logits and relaxes rejection only when strict verification provides minimal benefit. Importantly, the approach modifies only the verification rule and is fully compatible with existing target-coupled speculative decoding frameworks. Extensive experiments across model scales ranging from 8B to 235B demonstrate that our method delivers consistent and significant inference speedups over state-of-the-art baselines while preserving generation quality across diverse benchmarks. The code is available at https://github.com/5SSjw/MARS.

cs.LG↗
arXiv2026

Co-Evolution of Policy and Internal Reward for Language Agents

Large language model (LLM) agents learn by interacting with environments, but long-horizon training remains fundamentally bottlenecked by sparse and delayed rewards. Existing methods typically address this challenge through post-hoc credit assignment or external reward models, which provide limited guidance at inference time and often separate reward improvement from policy improvement. We propose Self-Guide, a self-generated internal reward for language agents that supports both inference-time guidance and training-time supervision. Specifically, the agent uses Self-Guide as a short self-guidance signal to steer the next action during inference, and converts the same signal into step-level internal reward for denser policy optimization during training. This creates a co-evolving loop: better policy produces better guidance, and better guidance further improves policy as internal reward. Across three agent benchmarks, inference-time self-guidance already yields clear gains, while jointly evolving policy and internal reward with GRPO brings further improvements (8\%) over baselines trained solely with environment reward. Overall, our results suggest that language agents can improve not only by collecting more experience, but also by learning to generate and refine their own internal reward during acting and learning.

cs.LG↗
arXiv2026

Teaching Large Reasoning Models Effective Reflection

Large Reasoning Models (LRMs) have recently shown impressive performance on complex reasoning tasks, often by engaging in self-reflective behaviors such as self-critique and backtracking. However, not all reflections are beneficial-many are superficial, offering little to no improvement over the original answer and incurring computation overhead. In this paper, we identify and address the problem of superficial reflection in LRMs. We first propose Self-Critique Fine-Tuning (SCFT), a training framework that enhances the model's reflective reasoning ability using only self-generated critiques. SCFT prompts models to critique their own outputs, filters high-quality critiques through rejection sampling, and fine-tunes the model using a critique-based objective. Building on this strong foundation, we further introduce Reinforcement Learning with Effective Reflection Rewards (RLERR). RLERR leverages the high-quality reflections initialized by SCFT to construct reward signals, guiding the model to internalize the self-correction process via reinforcement learning. Experiments on two challenging benchmarks, AIME2024 and AIME2025, show that SCFT and RLERR significantly improve both reasoning accuracy and reflection quality, outperforming state-of-the-art baselines. All data and codes are available at https://github.com/wanghanbinpanda/SCFT.

cs.AI↗