Search arXiv⌕ Search

arXiv subjects

Yichen Li

Publications and source records attributed to Yichen Li.

At least 37 records · Page 2Linked to original sources
arXiv2026

Ophiuchus: Incentivizing Tool-augmented "Think with Images" for Joint Medical Segmentation, Understanding and Reasoning

Recent medical MLLMs have made significant progress in generating step-by-step textual reasoning chains. However, they still struggle with complex clinical tasks that necessitate dynamic and iterative focusing on fine-grained visual regions. To close this gap, we introduce Ophiuchus, a versatile, tool-augmented framework that equips an MLLM to (i) decide when fine-grained visual evidence is needed, (ii) determine where to probe and ground within the medical image, and (iii) seamlessly weave the relevant sub-image content back into an interleaved, multimodal chain of thought for precise segmentation and diagnosis. Ophiuchus moves beyond mere tool-calling by tightly fusing the MLLM's inherent grounding and reasoning capabilities with external tools, enabling more accurate and trustworthy decisions. The core of our method is a three-stage training strategy: cold-start SFT for basic tool selection; self-reflection fine-tuning to strengthen decision revision; and agentic tool reinforcement learning to elicit sophisticated, expert-like diagnostic behaviors. Extensive experiments show that Ophiuchus consistently outperforms both closed-source and open-source SOTA methods across diverse medical benchmarks, including VQA, detection, and reasoning-based segmentation. Our project code is available at https://github.com/SII-zyj/Ophiuchus.

cs.AI↗
arXiv2026

Experience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory

Medical agent systems are increasingly expected to support interactive clinical decision making rather than only static question answering. In such settings, effective agents must reuse prior experience across evolving cases, yet existing memory mechanisms often retain raw historical traces that are redundant, noisy, and difficult to govern. More importantly, they rarely distinguish which memories are truly useful for future reasoning. This limits their ability to accumulate compact and reliable experience for long-horizon clinical reasoning. To close this gap, we propose SkeMex, a post-deployment self-evolution framework that improves medical agents through a skill-based memory without updating model weights. SkeMex distills informative interaction trajectories into structured skills that encode reusable procedural knowledge, and organizes them into a multi-branch repository spanning general, task-specific, and action-level experience. To determine which memories should be reused and retained, SkeMex estimates context-dependent utility from environment feedback and uses it to guide value-aware retrieval and repository governance. A closed-loop ``Read--Write--Assess--Govern" lifecycle further supports continual evolution by writing new skills, updating utilities, promoting useful memories, and removing harmful entries. Experiments across diverse clinical tasks show that SkeMex consistently outperforms representative memory-based agents in both offline and online settings. It also generalizes across model backbones and supports transferable skill memory. All data and code will be released publicly.

cs.AI↗
arXiv2026

Amortized Predictability-aware Training Framework for Time Series Forecasting and Classification

Time series data are prone to noise in various domains, and training samples may contain low-predictability patterns that deviate from the normal data distribution, leading to training instability or convergence to poor local minima. Therefore, mitigating the adverse effects of low-predictability samples is crucial for time series analysis tasks such as time series forecasting (TSF) and time series classification (TSC). While many deep learning models have achieved promising performance, few consider how to identify and penalize low-predictability samples to improve model performance from the training perspective. To fill this gap, we propose a general Amortized Predictability-aware Training Framework (APTF) for both TSF and TSC. APTF introduces two key designs that enable the model to focus on high-predictability samples while still learning appropriately from low-predictability ones: (i) a Hierarchical Predictability-aware Loss (HPL) that dynamically identifies low-predictability samples and progressively expands their loss penalty as training evolves, and (ii) an amortization model that mitigates predictability estimation errors caused by model bias, further enhancing HPL's effectiveness. The code is available at https://github.com/Meteor-Stars/APTF.

cs.LG↗
arXiv2026

Lost in the Flow with Code Talkers: Unveiling the Instruction-Tuning Tax of Large Language Models in Code Tasks

AI coding assistants have significantly improved developer productivity by automatically suggesting code that aligns with user intent, and many of these tools are now integrated directly into Integrated Development Environments (IDEs). Developers interact with code in two distinct cognitive modes: Flow and Command. While developers require tools that directly complete or infill code in unfinished programs during Flow mode, they also need tools that can comprehend intentions expressed as natural-language instructions and convert them into executable code in Command mode. Although instruction-tuned Large Language Models (LLMs) dominate many application scenarios due to their abilities to infer and fulfill developers' intents, it remains unclear whether the same paradigm is equally suitable for different code-related tasks. Therefore, it is necessary to understand how instruction tuning affects the feasibility of CodeLLMs as coding assistants. To fill this gap, we conduct the first empirical study that uncovers a key trade-off caused by instruction tuning across programming modes, which we term the Instruction-Tuning Tax. Our results show that instruction tuning is not a free lunch: although instruction-tuned models are more capable of following instructions and leveraging structured guidance, these gains often come at the cost of weaker infilling performance. We further extend our study through both qualitative and quantitative analyses, including manual failure categorization, behavioral metrics that capture generation fidelity, and intermediate-checkpoint evaluation throughout the tuning process. Summarizing our results into seven findings and four implications, our study offers a new perspective on the development of AI-powered coding tools and highlights the need to carefully balance instruction-following ability with effective code generation assistance.

cs.SE↗
arXiv2026

FingerEye: Learning Dexterous Manipulation with Continuous Vision-Tactile Sensing

Dexterous robotic manipulation requires perception that remains informative from pre-contact approach to contact initiation and post-contact control. We introduce FingerEye, a sensing and learning framework that strengthens robotic dexterity through continuous vision-tactile feedback throughout interaction. On the sensing side, FingerEye integrates binocular RGB cameras with a compliant contact interface to support perception both before and after contact. Before contact, the fingertip cameras provide close-range visual cues and implicit stereo for precise approach and object localization. After contact, marker-tracked deformation of the compliant ring provides a proxy for contact wrench sensing. On the learning side, we build real-and-sim infrastructure for data collection and evaluation, systematically study policy-interface designs for learning with multiple FingerEye sensors, and develop FingerEye Policy, which applies group-structured modality fusion to reduce modality shortcuts and better exploit distributed fingertip feedback. Across seven contact-sensitive task settings, FingerEye improves wrist-only policy by over 30 percentage points in mean success rate in both simulation and the real world.

cs.RO↗
arXiv2026

Test-Time Training for Zero-Resource Dense Retrieval Reranking

Dense retrievers excel at first-stage candidate generation but lack effective reranking in zero-resource settings. Existing approaches face a fundamental dilemma: cross-encoders deliver strong reranking quality but require costly supervised training and incur high latency, while unsupervised BM25 reranking consistently degrades dense retrieval performance on most of BEIR benchmarks. We propose DART (Dense Adaptive Reranking at Test-time), which resolves this dilemma by adapting the scoring function at inference time. For each query, the top-ranked documents serve as pseudo-positive examples and the bottom-ranked as pseudo-negative examples, providing noisy but readily available supervision to adapt a bilinear scoring matrix $W$ via a small number of gradient updates. We further introduce a confidence-weighted margin loss and a cross-query momentum buffer that warm-starts adaptation across queries. On six BEIR benchmarks, DART achieves a mean per-dataset relative NDCG@10 gain of +2.1% over the dense retrieval baseline with under 10ms additional latency per query, demonstrating a powerful capability for zero-shot performance enhancement and cross-domain generalization.

cs.IR↗
arXiv2026

TimeSage-MT: A Multi-Turn Benchmark for Evaluating Agentic Time Series Reasoning

Time series data inform critical decisions across many real-world domains. While large language model (LLM) agents can analyze data through natural language and tools, it remains unclear whether they can conduct reliable time series analysis across multi-turn conversations. Existing benchmarks focus on single-step tasks such as forecasting and anomaly detection, overlooking practical workflows where user goals evolve, agents must build on prior analyses, and conclusions emerge from accumulated evidence. In this work, we introduce TimeSage-MT, a multi-turn benchmark for agentic time series reasoning with 240 tasks and 2,680 dialogue turns across 8 real-world domains, spanning basic exploration to decision-oriented analysis. TimeSage-MT is built through a reproducible pipeline that converts real-world time series data into multi-turn conversations with verifiable answers. It provides a unified evaluation protocol and public leaderboard for comparing time series agentic systems. To demonstrate the benchmark's utility, we evaluate frontier LLMs alongside TimeSage, a novel structured agent equipped with a comprehensive time series skill library. The results show sharp performance drops on decision-oriented tasks, driven by failures in memory, uncertainty handling, and domain-based decision making. TimeSage-MT exposes critical gaps in current agentic reasoning and provides a rigorous foundation for future development.

cs.CL↗
arXiv2026

CoCoReviewBench: A Completeness- and Correctness-Oriented Benchmark for AI Reviewers

Despite the rapid development of AI reviewers, evaluating such systems remains challenging: metrics favor overlap with human reviews over correctness. However, since human reviews often cover only a subset of salient issues and sometimes contain mistakes, they are unreliable as gold references. To address this, we build category-specific benchmark subsets and skip evaluation when the corresponding human reviews are missing to strengthen Completeness. We also leverage reviewer--author--meta-review discussions as expert annotations and filter unreliable reviews accordingly to strengthen Correctness. Finally, we introduce CoCoReviewBench, which curates 3,900 papers from ICLR and NeurIPS to enable reliable and fine-grained evaluation of AI reviewers. Analysis shows that AI reviewers remain limited in correctness and are prone to hallucinations, and highlights reasoning models as more effective reviewers, motivating further directions for improving AI reviewers. Benchmarks and models are available at https://github.com/hexuandeng/CoCoReviewBench.

cs.CL↗
arXiv2026

Implementation of a Low-Temperature Monitoring and Alarm System for the Taishan Neutrino Experiment

The Taishan Antineutrino Observatory (TAO) is a near-site experiment for the Jiangmen Underground Neutrino Observatory (JUNO). Its primary goal is to provide a precise reference reactor antineutrino energy spectrum, thereby eliminating the model dependence in reactor neutrino spectrum predictions and enhancing the sensitivity of the neutrino mass ordering measurement. To ensure accurate data acquisition and safe operation of the TAO experiment, a low-temperature monitoring and alarm system has been developed. Built on the Experimental Physics and Industrial Control System (EPICS) framework, the system employs PT100 platinum resistance thermometers embedded in the detector to monitor the temperature of the liquid scintillator. Real-time temperature data are acquired, enabling comprehensive thermal monitoring. The alarm program adopts a trigger-based mechanism with multi-level thresholds, providing instant alerts to operators when the temperature deviates from the safe range. The system has been operating stably for six months, accumulating over one thousand alarm records, and has proven effective in ensuring the safe and stable operation of the experiment.

physics.ins-det↗
arXiv2026

Reference-based Category Discovery: Unsupervised Object Detection with Category Awareness

Traditional one-shot detection methods have addressed the closed-set problem in object detection, but the high cost of data annotation remains a critical challenge. General unsupervised methods generate pseudo boxes without category labels, thus failing to achieve category-aware classification. To overcome these limitations, we propose Reference-based Category Discovery (RefCD), an unsupervised detector that enables category-aware\footnotemark[1] detection without any manually annotated labels. It leverages feature similarity between predicted objects and unlabeled reference images. Unlike previous unsupervised methods that lack category guidance and one-shot methods which require labeled data, RefCD introduces a carefully designed feature similarity loss to explicitly guide the learning of potential category-specific features. Additionally, RefCD supports category-agnostic detection without reference images, serving as a unified framework. Comprehensive quantitative and qualitative analysis of category-aware and category-agnostic detection results demonstrates its effectiveness, and RefCD can learn category information in an unsupervised paradigm even without category labels.

cs.CV↗
arXiv2026

Industrial Deposition of Wavelength-Shifting Films for Liquid Argon Photon Detection Systems

The Deep Underground Neutrino Experiment (DUNE) Phase-II Far Detector is considering an approximately 2000\,m$^2$ photon detection system to achieve a target mean light yield of 180\,PE/MeV. Meeting this requirement demands scalable, cost-effective, and high-quality wavelength-shifter (WLS) coatings capable of converting 127\,nm liquid-argon scintillation light into visible photons with controlled and reproducible optical performance. We report on the successful realization of an industrial physical vapor deposition (PVD) process for \textit{p}-terphenyl (pTP) coatings, adapted from vacuum deposition techniques developed for OLED display manufacturing, to produce uniform WLS layers on large-area inorganic substrates, a task traditionally challenged by adhesion and uniformity issues at organic--inorganic interfaces. Surface characterization by profilometry and spectroscopic measurements demonstrates edge-region thickness variation below 10\% and emission spectra consistent with high-quality pTP reference samples. The industrial process demonstrates reproducibility, scalability, and significantly reduced production time compared to laboratory-based methods, while maintaining optical characteristics consistent with established pTP reference samples. These results establish a viable pathway for mass production of high-performance pTP coatings for DUNE FD3 and future neutrino experiments, from a coating manufacturing and process standpoint. Detector-level performance validation, including quantitative VUV conversion efficiency measurements at 127\,nm, is identified as future work.

physics.ins-det↗
arXiv2026

A 260-Liter Test Stand for Liquid Argon R&D

We describe the design and performance of a 260-liter liquid argon (LAr) cryogenic test stand for liquid argon detector research and development at BNL. The system uses gas-phase argon purification with continuous pump-free circulation, in which boil-off argon gas is purified, recondensed, and returned to the cryostat by gravity without a mechanical recirculation pump; it also incorporates an upgraded condenser that increases the effective thermal contact area by a factor of 13 relative to the previously developed 20-liter system reported perviously. A liquid argon purity monitor is installed to measure the electron lifetime directly in LAr, enabling quantitative characterization of charge attenuation due to electronegative impurities. Under the operating conditions reported here, the demonstrated electron lifetime is 0.5 ms. The system is designed to enable rapid iteration of detector components in complete operational cycles, including pump-down, leak verification, cryogenic fill, stable operation, and warm-up, which can be completed within 7 days. Such a fast turnaround time, together with the medium-scale liquid volume and direct purity diagnostics, makes the facility well suited for testing and refining detector designs in support of large liquid argon time projection chamber (LArTPC) experiments.

physics.ins-det↗
arXiv2026

TypeScript Repository Indexing for Code Agent Retrieval

Graph-based code indexing can improve context retrieval for LLM-based code agents by preserving call chains and dependency relationships that keyword search and similarity retrieval often miss. ABCoder is an open-source framework that parses codebases into a function-level code index called UniAST. Its existing parsers combine lightweight AST parsers for syntactic analysis with language servers for semantic resolution, but because LSP-based resolution requires a JSON-RPC call for each symbol lookup, these per-symbol calls become a bottleneck on large TypeScript repositories. We present abcoder-ts-parser, a TypeScript parser built on the TypeScript Compiler API that works directly with the compiler's AST, semantic information, and module resolution logic. We evaluate the parser on three open-source TypeScript projects with up to 1.2 million lines of code and find that it produces reliable indexes significantly more efficiently than the existing architecture. For a live demonstration, watch: https://youtu.be/ryssr7ouvdE

cs.SE↗
arXiv2026

How Far Are Video Models from True Multimodal Reasoning?

Despite remarkable progress toward general-purpose video models, a critical question remains unanswered: how far are these models from achieving true multimodal reasoning? Existing benchmarks fail to address this question rigorously, as they remain constrained by straightforward task designs and fragmented evaluation metrics that neglect complex multimodal reasoning. To bridge this gap, we introduce CLVG-Bench, an evaluation framework designed to probe video models' zero-shot reasoning capabilities via Context Learning in Video Generation. CLVG-Bench comprises more than 1,000 high-quality, manually annotated metadata across 6 categories and 47 subcategories, covering complex scenarios including physical simulation, logical reasoning, and interactive contexts. To enable rigorous and scalable assessment, we further propose an Adaptive Video Evaluator (AVE) that aligns with human expert perception using minimal annotations, delivering interpretable textual feedback across diverse video context tasks. Extensive experiments reveal a striking answer to our central question: while state-of-the-art (SOTA) video models, such as Seedance 2.0, demonstrate competence on certain understanding and reasoning subtasks, they fall substantially short with logically grounded and interactive generation tasks (achieving success rates <25% and ~0%, respectively), exposing multimodal reasoning and physical grounding as critical bottlenecks. By systematically quantifying these limitations, the proposed method provides actionable feedbacks and a clear roadmap toward truly robust, general-purpose video models. CLVG-Bench and code are released here.

cs.CV↗
arXiv2026

Cascaded Code Editing: Large-Small Model Collaboration for Effective and Efficient Code Editing

Code editing constitutes a fundamental practice in software development, wherein developers modify existing codebases according to natural language requirements. Accurate code editing necessitates a comprehensive understanding of both the existing codebase and the modification requirements. Although large language models (LLMs) have demonstrated promising performance in code editing tasks, they suffer from substantial inefficiency by generating entire modified files that largely consist of unchanged code. While smaller models could potentially address this inefficiency, they typically lack the capacity to effectively comprehend long code contexts required for accurate editing. To ensure both effectiveness and efficiency, we propose to decompose code editing into a two-stage cascade: \textbf{edit sketch generation}, wherein a large model first produces concise sketches representing the requisite modifications (the more challenging phase), and \textbf{edit sketch application}, wherein a smaller model integrates these sketches into the original code to produce the final output edited code (the simpler phase). This cascaded design reduces the number of tokens generated by the large model, as the majority of the output is handled by the smaller, more efficient model, thereby enhancing overall efficiency. However, the effectiveness of this approach is constrained by current small models' limited capabilities in handling long-context scenarios and cross-file dependencies, which are essential for accurate sketch application in real-world codebases. To address these limitations and enhance smaller models' sketch application capabilities, ...

cs.SE↗
arXiv2026

Lishu: A Real-Source Research Workbench for Elite Business Journal Search, Analysis, and Writing Support

This paper presents Lishu, a deployable web artifact for searching, monitoring, and interpreting literature from elite business and management journals. The system integrates the UTD-24 and Financial Times 50 (FT50) journal pools and combines Crossref, OpenAlex, Unpaywall, and optional CORE enrichment to support a broader research workflow than article retrieval alone. In the current implementation, users can search across curated journal pools, apply multi-journal filters, preview open full-text excerpts when available, generate citations and exports, inspect topic and affiliation structure, produce review drafts, simulate virtual peer review, and assemble grant-oriented research narratives. Unlike static journal directories or general-purpose academic search engines, the artifact is explicitly scoped to high-status management outlets and is designed to support sensemaking tasks that matter to researchers, doctoral students, and lab managers: identifying recent work, surfacing topical concentration, comparing themes, and converting search output into actionable research material. Architecturally, the system emphasizes source transparency, modularity, and low-cost public deployability through a lightweight Node.js service layer, a multi-page client interface, optional large-language-model enhancement for interpretation and writing support, and a free-tier persistence path through Supabase. The paper contributes both a functioning design artifact and an extensible architectural pattern for journal-pool-specific scholarly discovery and writing support, with implications for digital research infrastructure in information systems and business scholarship.

cs.DL↗
arXiv2026

Understanding Secret Leakage Risks in Code LLMs: A Tokenization Perspective

Code secrets are sensitive assets for software developers, and their leakage poses significant cybersecurity risks. While the rapid development of AI code assistants powered by Code Large Language Models (CLLMs), CLLMs are shown to inadvertently leak such secrets due to a notorious memorization phenomenon. This study first reveals that Byte-Pair Encoding (BPE) tokenization leads to unexpected behavior of secret memorization, which we term as \textit{gibberish bias}. Specifically, we identified that some secrets are among the easiest for CLLMs to memorize. These secrets yield high character-level entropy, but low token-level entropy. Then, this paper supports the biased claim with numerical data. We identified that the roots of the bias are the token distribution shift between the CLLM training data and the secret data. We further discuss how gibberish bias manifests under the ``larger vocabulary'' trend. To conclude the paper, we discuss potential mitigation strategies and the broader implications on current tokenizer design.

cs.CR↗
arXiv2026

CHIPS: Efficient CLIP Adaptation via Curvature-aware Hybrid Influence-based Data Selection

Adapting CLIP to vertical domains is typically approached by novel fine-tuning strategies or by continual pre-training (CPT) on large domain-specific datasets. Yet, data itself remains an underexplored factor in this process. We revisit this task from a data-centric perspective: Can effective data selection substitute for large-scale datasets in CPT? We introduce CHIPS (Curvature-aware Hybrid Influence in Projection Subspace), which assigns each image-text pair a utility score that integrates three complementary factors aligned with three goals: faithfulness via a curvature-aware and Newton-style alignment computed in CLIP's end-point subspace; scalability via an InfoNCE-aware curvature estimator with Johnson-Lindenstrauss (JL) sketching; and retention via a selection-aware relevance weight combined with learnability to balance target adaptation against general-domain preservation. We justify this design theoretically by proving a lower-bound guarantee on the proxy's correlation with full-parameter alignment and by characterizing the bias-variance trade-offs introduced by curvature mixing and JL sketching. We evaluate CHIPS empirically across various settings: 1) CHIPS attains state-of-the-art performance among selection baselines on 17 medical benchmarks, matches full-dataset CPT with 30% of the data, and outperforms half-dataset CPT using only 10%; 2) on 31 general-domain benchmarks, CHIPS yields the least performance drop under all retention ratios.

cs.LG↗