Search arXiv⌕ Search

arXiv subjects

Hong Chen

Publications and source records attributed to Hong Chen.

At least 19 recordsLinked to original sources
arXiv2026

EG-GRPO: Evidence-Guided Step-level Credit Assignment for Deep Search Agents

LLM-based search agents demonstrate strong capabilities in long-horizon information seeking and reasoning, typically trained via reinforcement learning. However, existing approaches predominantly rely on trajectory-level outcome rewards, providing limited supervision for complex search and reasoning processes. Although step-level credit assignment offers a promising solution, obtaining reliable and task-grounded supervision for open-ended search remains challenging. In this paper, we propose Evidence-Guided Group Relative Policy Optimization (EG-GRPO), a framework that derives step-level supervision from the evidence annotations available in synthetic deep search data. EG-GRPO identifies the first supporting citation for each required evidence claim and uses the newly acquired evidence to construct step-level process rewards. By representing the accumulated evidence as an evolving knowledge state, EG-GRPO further performs state-conditioned relative advantage estimation and combines the resulting step-level signal with the standard trajectory-level outcome advantage. Experiments on five challenging deep search benchmarks show that EG-GRPO generally outperforms standard GRPO and existing reward-enhanced and step-level RL baselines across different model scales. These results demonstrate that task-grounded evidence annotations can provide effective process supervision for credit assignment in deep search RL.

cs.LG↗
arXiv2026

Selm: Self-Evolving Library Management for Multi-Skill Agent Optimization

Large language model (LLM) agents increasingly use skills as a lightweight way to acquire task-specific knowledge. Recent work enables skills to evolve from task trajectories. However, these methods often lack systematic management of evolved skills, making it difficult for agents to reuse and accumulate skills across domains, thus limiting performance on subsequent tasks.To reliably accumulate, reuse, and maintain useful skills across tasks, we present SELM (Self-Evolving Library Management), a framework that formulates skill evolution as library-level management. SELM jointly manages skill retrieval, updating, and lifecycle by maintaining skill confidence and utility signals. It introduces utility-aware retrieval to select useful skills, a Map-Reduce framework to generate targeted updates from task-specific experience, and lifecycle controls to retain or remove skills based on their effectiveness. These mechanisms enable the skill library to continuously accumulate useful experience while preserving and pruning skills across tasks, improving agent performance on complex tasks across diverse domains. We evaluate SELM on four agent benchmarks covering diverse tasks and execution environments, and SELM consistently outperforms in most settings.

cs.AI↗
arXiv2026

Interplay between high-entropy effects and the superionic transition in fluorite oxides: A liquid-like perspective

Minor actinide (MA)-bearing mixed oxide (MOX), a prominent high-entropy ceramic (HEC), is a promising candidate for advanced fast reactor fuels. Using a hybrid Monte Carlo/Molecular Dynamics (MC/MD) approach, we reveal that Np and Am segregation establishes a distinct short-range order (SRO) that induces severe lattice distortion. We uncover a competitive mechanism between localized high-entropy blocking effects and collective superionic dynamics, where the anionic kinetic and structural evolution within the 2000-2800 K range is partitioned and governed by the Tammann crossover temperature and superionic transition temperature. In the low-temperature regime, severe lattice distortions induced by the SRO structures create localized electrostatic and strain-coupled potential wells, which spatially confine the O-ions and lead to their sluggish diffusion. The system exhibits increased dynamical heterogeneity as thermal activation alleviates these structural constraints. Crucially, these localized traps spontaneously transform into a highly interconnected, string-like polyhedral topological network. Driven by the maximization of configurational entropy, this structural change creates low-viscosity transport pathways that trigger confinement-enhanced diffusion. Consequently, a dramatic kinetic inversion occurs, where the O-ion mobility of SRO model surpasses that of random solid solution (RSS) model via an entropy-driven compensation mechanism. Finally, through Gibbs free energy minimization and defect analysis, structural and thermodynamic origins of the superionic transition in HEC are elucidated. This study unveils how local chemical ordering regulates ion transport under extreme conditions, providing theoretical support for the design of high-performance nuclear fuels with both superior radiation tolerance and optimized mass transfer capabilities.

cond-mat.mtrl-sci↗
arXiv2026

Multi-LLM Orchestration for High-Quality Code Generation: Exploiting Complementary Model Strengths

Large Language Models (LLMs) are central to automated code generation, yet most approaches use one model at every stage, from writing code to repairing and optimizing it. LLMs have complementary strengths: their failures do not fully overlap across programming languages, algorithmic categories, and development stages. For example, in Python, the model with the highest code-generation accuracy ranks last at code repair. Using one model to generate code, another to repair it, and a third to refine it can therefore achieve higher correctness and better execution performance than a single model throughout. We propose PerfOrch, a multi-LLM multi-agent system with four agents for categorization, generation, debugging, and refinement. The generation, debugging, and refinement agents select models through a Memory module that ranks candidate LLMs by their measured performance for each stage, programming language, and algorithmic category. We evaluate PerfOrch with five candidate LLMs on 4,640 problems from four benchmarks covering five programming languages (HumanEval-X, EffiBench-X, BigCodeBench, and APPS) against single-model direct generation, single-model generate-debug-refine pipelines (PerfCodeGen), and the state-of-the-art multi-agent systems MapCoder and CodeSIM. PerfOrch achieves the highest pass@1 in all twelve benchmark-language settings, exceeding the strongest baseline by an average of 8.19 percentage points, and reduces execution time for an average of 75.03% of solved problems with an average speedup of 1.18x. The Memory rankings remain effective on three unseen benchmarks without re-profiling. Sequential refinement matches exhaustive execution-time refinement coverage with 66% fewer refinement-stage tokens (57% for average-memory optimization). The complementary strengths persist after all five LLMs are replaced with their current-generation successors.

cs.SE↗
arXiv2026

D$^{2}$-VPR: A Parameter-efficient Visual-foundation-model-based Visual Place Recognition Method via Knowledge Distillation and Deformable Aggregation

Visual Place Recognition (VPR) aims to determine the geographic location of a query image by retrieving its most visually similar counterpart from a geo-tagged reference database. Recently, the emergence of the powerful visual foundation model, DINOv2, trained in a self-supervised manner on massive datasets, has significantly improved VPR performance. This improvement stems from DINOv2's exceptional feature generalization capabilities but is often accompanied by increased model complexity and computational overhead that impede deployment on resource-constrained devices. To address this challenge, we propose $D^{2}$-VPR, a $D$istillation- and $D$eformable-based framework that retains the strong feature extraction capabilities of visual foundation models while significantly reducing model parameters and achieving a more favorable performance-efficiency trade-off. Specifically, first, we employ a two-stage training strategy that integrates knowledge distillation and fine-tuning. Additionally, we introduce a Distillation Recovery Module (DRM) to better align the feature spaces between the teacher and student models, thereby minimizing knowledge transfer losses to the greatest extent possible. Second, we design a Top-Down-attention-based Deformable Aggregator (TDDA) that leverages global semantic features to dynamically and adaptively adjust the Regions of Interest (ROI) used for aggregation, thereby improving adaptability to irregular structures. Extensive experiments demonstrate that our method achieves competitive performance compared to state-of-the-art approaches. Meanwhile, it reduces the parameter count by approximately 64.2% (compared to CricaVPR).Code is available at https://github.com/tony19980810/D2VPR.

cs.CV↗
arXiv2026

Small Frequency Corrections Can Change What Survives KV Cache Compression

Compressing a key-value cache before its next question is known requires choosing what to retain without knowing which evidence will matter. Value energy measures entry strength but does not distinguish isolated keys from those with many similar neighbors. We introduce TwinKV, a training-free method that discounts value energy by nonlocal post-RoPE key frequency. Prefix attention allocates head capacities, while retained entries preserve their original keys and values under an exact storage budget. Across four language models, TwinKV exceeds five evaluated compressed baselines in mean score on LongBench, LooGLE, and RULER at 50\% KV removal. Component controls isolate the frequency contribution. On Llama-3.2-1B RULER at 75\% removal, normalized frequency weights average 0.95, yet change 7\% of nonprotected retained positions and improve value-only retention by about 5.5 points under both uniform and adapted capacities. Permuting the weights within heads weakens this gain. These results show that modest frequency corrections can change retention and answering outcomes, with effects that depend on the model and task.

cs.CL↗
arXiv2026

When Keywords Drop but Classifiers Hold: Soft Refusals under KV Cache Compression

KV cache compression is widely used for long context LLM inference under memory constraints, while deployed systems typically score refusals after generation with keyword filters or learned classifiers. Such monitors are intended to indicate whether a model declined a harmful request under the serving regime actually used. However, it remains unclear whether matched compression that preserves task accuracy also preserves agreement between lightweight lexical monitors and stronger refusal classifiers. We study this with a paired protocol on n=200 harmful prompts with a long filler context: each prompt is answered once under full retention and once under matched eviction after a shared prefill, and the same replies are scored by keyword heuristics, the HarmBench Llama-2-13B classifier, an auxiliary LLM judge, and humans on disagreements. On Qwen2.5-3B, keyword refusal falls from 98.0% to 80.5% (McNemar p~1e-8) while classifier refusal stays near ceiling (99.0%-99.5%) and MMLU accuracy is unchanged (50.0%); human labels predominantly follow the classifier, consistent with soft refusals. The gap is not universal and weakens under short fillers and paired SnapKV, so safety auditing under compression should rely on several judges matched to the serving context rather than on keyword rates alone.

cs.LG↗
arXiv2026

Vision-Guided Text Prompt Tuning for Multimodal Sentiment Analysis

Multimodal sentiment analysis requires effective modeling of both verbal semantics and non-verbal affective cues. A central challenge is to calibrate text-centered sentiment understanding with visual facial evidence in a controlled, adaptive, and parameter-efficient manner. Text usually serves as the semantic anchor, whereas visual cues provide complementary evidence for ambiguous or implicit expressions; however, indiscriminate fusion may introduce visual noise and distort textual semantics. Moreover, fully fine-tuning large visual and textual encoders is costly and prone to overfitting on limited and scenario-dependent MSA benchmarks. To address these issues, we propose Vision-Guided Text Prompt Tuning (VG-TPT), which formulates visual-text sentiment modeling as controllable visual calibration of frozen text representations. VG-TPT injects visual affective cues into a frozen BERT encoder through layer-wise adaptive prompts, rather than relying on late-stage feature fusion or full backbone tuning. A co-guided router composes prompts from a trainable prompt bank according to both the evolving text state and the visual guidance feature, enabling sample-specific and layer-specific modulation. Experiments on CMU-MOSEI and CMU-MOSI show that VG-TPT consistently improves over text-only baselines and achieves competitive or superior performance compared with several full-modality methods, while updating only 2.4M trainable parameters. The code is available at https://github.com/ma-tubu/VG-TPT.

cs.MM↗
arXiv2026

SemTrace: Source-Grounded Semantic Signatures for Tracing LLM Exposure to Protected Documents

Large language models are increasingly used to read documents and produce downstream text, creating a provenance problem when the document owner cannot control or inspect the model that performs the generation. We introduce SemTrace, a source-grounded semantic watermark for detecting whether a generated review was influenced by a known protected manuscript copy. Rather than biasing token probabilities or imposing surface-form patterns, SemTrace constructs a document-specific binary signature from factual propositions that are directly supported by the manuscript itself. A protected PDF invisibly carries a content contract that selects one fact from each binary pair and asks an instruction-following reviewer to express those facts in fixed review slots without changing its independent evaluation. A frozen natural language inference model then decodes the resulting semantic evidence with explicit erasures and scores the recovered bits against the codeword assigned to that copy. This design targets model-agnostic, assigned-copy exposure detection while keeping the watermark semantically tied to the source document.

cs.CL↗
arXiv2026

NestedKV: Nested Memory Routing for Long-Context KV Cache Compression

Long-context language models are limited by the memory footprint of the key-value (KV) cache. Existing training-free KV compression methods usually rank tokens by one importance signal -- attention, recency, layer-wise allocation, or key distinctiveness -- which becomes brittle when useful context is globally distinctive, locally episodic, or immediately relevant. We introduce NestedKV, a key-only KV cache compression method inspired by the Continuum Memory System in Nested Learning. NestedKV maintains global, block-level, and sliding-window key anchors, scores tokens by multi-time-scale cosine anomaly, and combines the resulting rankings with a training-free outer learner using head-adaptive mixing and surprise-gated token routing. The score is paired with adaptive per-head budgets and requires no training or LLM modification. Across RULER (4k--32k), LooGLE, LongBench, LongBench-E, InfiniteBench, and MMLU-Pro on Qwen3 and Llama-3.2 models, NestedKV is strongest when the retained cache is small. On Qwen3-4B, it improves over KeyDiff by up to 19.10 points on RULER and 19.29 on LongBench at $r=0.75$; at $r=0.95$, it retains 37.32 on LongBench versus 17.55 for KeyDiff.

cs.CL↗
arXiv2026

PANDA: A Matrix-Free Differentiable NMPC Solver via Proximal Averaged Quasi-Newton with Adaptive Linesearch Algorithm

Differentiable nonlinear model predictive control (NMPC) provides a principled way to embed optimal control structure into end-to-end learning paradigms, but its practical use is often limited by the computational and memory costs of both forward optimization and backward sensitivity propagation. This brief proposes PANDA, a matrix-free solver for differentiable NMPC. In the forward pass, PANDA combines proximal-gradient iterations with quasi-Newton acceleration and introduces an adaptive stepsize enlargement mechanism to mitigate the conservativeness of monotone stepsize reduction. The resulting stepsize behavior and its effect on local convergence are theoretically analyzed. In the backward pass, PANDA performs implicit differentiation from the residual equation and computes adjoint sensitivities using Krylov-subspace iterative methods together with automatic-differentiation-based Matrix-Vector product operators, thereby avoiding explicit Hessian and Jacobian construction. The method is evaluated on a nonconvex trailer NMPC problem embedded in an imitation learning task. The results show that PANDA achieves much faster forward and backward computation and lower memory overhead than representative differentiable optimization solvers, while maintaining effective imitation learning performance.

eess.SY↗
arXiv2026

ACTS-SQL: Agentic and Critic-Oriented Tree-Structured SQL Correctness with Large Language Models

Large Language Models (LLMs) have been increasingly adopted in Text-to-SQL systems, yet SQL errors remain a major obstacle in real-world Text-to-SQL inference pipelines. Existing SQL correction approaches either rely on large-scale, high-quality training data with substantial overhead, or adopt single-path agentic workflows that are brittle to early mistakes and prone to error propagation. To develop a practical SQL correctness system for industrial scenarios, we present a training-free framework that formulates SQL correction as a plan-guided, tree-structured debugging process. By maintaining multiple correction strategies and enabling backtracking, the framework mitigates error accumulation during iterative refinement. We further integrate execution-based verification and clause-level diagnostic tools to support strategy pruning and precise error localization. We evaluate the system on the BIRD-Critic benchmark and observe consistent accuracy gains over strong LLM backbones and representative agent-based baselines, achieving a 9.42% improvement over the previous state-of-the-art method. The framework is also deployed in the Torch Log Service (TLS) of Volcano Engine to support an online Text-to-TLS API. In production, it improves execution accuracy from 36.77% to 53.61% on real user queries with a representative strong LLM backbone (GPT-5). These results demonstrate the effectiveness and stability of our approach in real-world deployments.

cs.AI↗
arXiv2026

Integrated Energy System Optimal Operation in Coal District With Hydrogen Heavy Trucks

The coal industry contributes significantly to the social economy, but the emission of greenhouse gases puts huge pressure on the environment during mining, transportation, and power generation. In the integrated energy system (IES), current research on power-to-gas (P2G) technology mainly focuses on injecting hydrogen generated from renewable-energy-powered electrolysis into natural gas pipelines. However, hydrogen injection may cause irreversible hydrogen embrittlement of pipelines. In this paper, sufficient hydrogen is produced through both P2G technology and coal-to-hydrogen (C2H) based on coal gasification, considering the typical scenario where coal-producing regions are rich in both coal and renewable energy resources. To transport mined coal, hydrogen heavy-duty trucks provide a promising solution by consuming hydrogen in a timely manner, thereby avoiding potentially dangerous hydrogen injection into pipelines and reducing the need for relatively expensive hydrogen storage. An optimal scheduling model for an electric-gas IES is proposed based on second-order cone programming (SOCP). The proposed model innovatively studies a closed industrial loop including coal mining, hydrogen production, coal transportation by hydrogen trucks, and integrated energy systems to enhance renewable energy utilization and coordinate multiple energy carriers. Finally, a case study based on the IEEE 30-bus power system and the Belgium 24-node natural gas network is conducted. The results show that the proposed hydrogen production technology reduces typical daily operating costs by 7.7%. Under China's carbon emissions trading system, the operating costs of hydrogen heavy-duty trucks are reduced by 0.95% and 4.68% compared with electric vehicles and diesel trucks, respectively. Under Europe's stricter carbon emissions trading system, ...

eess.SY↗
arXiv2026

DocTrace: Towards Traceable Long Document VQA via Hierarchical Evidence Graph Reasoning

Long Document Visual Question Answering (LongDocVQA) requires Multimodal Large Language Models (MLLMs) to locate, integrate, and reason over heterogeneous document elements distributed across multiple pages. Existing approaches, including end-to-end MLLMs, retrieval-augmented generation (RAG) pipelines, and document agents, often lack explicit mechanisms to represent and verify how grounded evidence is progressively composed during reasoning, limiting both answer accuracy and traceability. In this paper, we cast LongDocVQA as an explicit evidence graph reasoning problem rather than implicit answer prediction. To this end, we propose DocTrace, a hierarchical framework that progressively performs evidence localization, structured document parsing, and evidence graph reasoning to enable explicit evidence provenance. To effectively learn these capabilities, we develop a two-stage training framework: joint Supervised Fine-Tuning (SFT) first initializes evidence localization and graph reasoning abilities, followed by task-specific Group Relative Policy Optimization (GRPO) with dedicated rewards to further optimize these capabilities. Extensive experiments on MMLongBench-Doc, LongDocURL, and SlideVQA demonstrate that DocTrace consistently outperforms both existing open-source baselines and proprietary MLLMs. Compared with the Qwen3-VL-8B-Instruct backbone, DocTrace achieves absolute improvements of 14.4, 11.3, and 11.7 points on the three benchmarks, respectively. Beyond competitive performance, DocTrace constructs traceable evidence graphs with explicit node-level provenance, enabling transparent and verifiable reasoning for long document understanding.

cs.AI↗
arXiv2026

Seen, Said, or Forgotten? A Causal Audit of Visual KV Memory Across Dialog Turns

Stateful multimodal assistants encode an image once but may answer questions about it many turns later. Attention-guided visual-KV eviction assumes that evidence irrelevant now will remain dispensable, although future questions are unknown. We ask when a visual fact is actually safe to forget and introduce the Causal Visual Memory Audit (CVMA), a paired single-prefill framework that tests what later answers lose when a visual region, the whole image, or prior assistant text becomes unavailable. On VisDial and ConvBench, current attention can rank future-useful regions worse than random even though a diagnostic marginal-utility control shows substantial selection headroom. Aggregate scores hide this failure when later turns do not need vision; controlled and stock-generated histories reveal a second escape route, in which assistant-text KV replaces image KV for facts already stated but not reliably for unstated facts. In the tested stacks, safe forgetting is supported by low future visual dependence or fact-specific verbalization---not by low current attention.

cs.CV↗
arXiv2026

Beyond GDPR: Examining Disclosure Gaps in Mobile AR Privacy Policies under U.S. State Privacy Laws

Mobile Augmented Reality (MAR) apps can collect and process highly sensitive data such as spatial maps and biometrics, yet their privacy policies remain largely understudied. Prior audits of app privacy policies have typically focused on a single legal framework, such as the GDPR. Meanwhile, 20 U.S. states have comprehensive privacy laws in effect, creating a fragmented and rapidly evolving set of privacy policy obligations. To date, no study has systematically audited privacy policies against this emerging body of state-level legislation. In this paper, we present the first large-scale audit of MAR privacy policies under U.S. state privacy laws. We construct a dataset covering the MAR ecosystem, including 8,013 Google Play MAR app metadata records worldwide, and a U.S.-based subset with 6,620 APKs and 6,426 privacy policy files. We further derive an auditable disclosure taxonomy with 5 baseline requirements, 10 triggered requirements, and 4 logic chains, and build a validated four-stage automated pipeline that produces traceable, evidence-grounded disclosure judgments. Our audit reveals widespread disclosure gaps: 44.62\% of audited policies exhibit severe disclosure omissions, with each missing more than eight requirements, and four privacy-policy requirements have violation rates above 90\%. These findings suggest that MAR privacy disclosures are not keeping pace with the growing complexity of U.S. state privacy regulation. We release our dataset, taxonomy, and auditing pipeline to support future research on scalable privacy compliance auditing.

cs.CR↗
arXiv2026

IDSTune: A Multi-Agent Collaborative Framework for Integrated Database System Tuning

Database tuning is critical for achieving high performance in modern database management systems (DBMSs). Existing methods typically optimize a single component---knobs, indexes, or materialized views---without accounting for their interdependencies. This limitation arises because these components require different tuning strategies and are difficult to integrate within a unified framework. As a result, directly extending a method to multiple components or simply combining separate methods often fails to capture cross-component collaboration and shared tuning signals. Moreover, existing methods are insufficient for handling diverse workloads, evolving data, and dynamic query patterns. To address these limitations, we propose IDSTune, an integrated tuning framework that jointly optimizes multiple configuration components through LLM-driven multi-agent collaboration. IDSTune operates in two phases: (i) workload compression, which extracts and selects task-relevant features, and (ii) configuration recommendation, where specialized agents collaboratively generate and refine configurations for knobs, indexes, and materialized views under the supervision of a centralized coordinator. By incorporating feedback and external knowledge retrieval, IDSTune achieves efficient and globally consistent tuning. Extensive experiments show that IDSTune achieves up to 38% performance improvement and 57% faster tuning, with strong adaptability across diverse scenarios.

cs.DB↗
arXiv2026

Should Missing Modalities Always Be Necessary to Repair for Multi-modal Sentiment Analysis?

Existing methods for multimodal sentiment analysis (MSA) under missing modalities usually follow a repair-first paradigm. We revisit this assumption and ask: \emph{should every missing modality be repaired?} A per-sample oracle analysis shows the answer is not always: full-modality input is optimal for only a small fraction of samples, and every modality subset is preferred by some samples. These results suggest that adding or repairing modalities may not always improve prediction, and that the utility of each modality is sample-dependent. Building on this finding, we propose \textbf{S}ufficiency-\textbf{I}nformed \textbf{E}vidential \textbf{V}al\textbf{vE} (\textbf{SIEVE}) that turns ``whether to repair'' into an explicit, learnable decision at the sample level. SIEVE compares a direct prediction branch with a repair branch, derives an empirical sufficiency signal from their per-sample loss gap, and routes each input through an evidential gate that jointly models sufficiency and its epistemic uncertainty. SIEVE is repair-agnostic: it operates as a plug-and-play decision on top of any explicit or implicit repair module, without modifying its internal design. Experiments on CMU-MOSI and IEMOCAP show that SIEVE consistently improves representative repair backbones across evaluated missing rates, and approaches the per-sample dual-branch achievable optimum.

cs.CL↗