Searcharxiv⌕ Search

arXiv subjects

Yifan Zhang

Publications and source records attributed to Yifan Zhang.

At least 109 records · Page 6Linked to original sources
arXiv2026

LLM-Based Scientific Equation Discovery via Physics-Informed Token-Regularized Policy Optimization

Symbolic regression aims to distill mathematical equations from observational data. Recent approaches have successfully leveraged Large Language Models (LLMs) to generate equation hypotheses, capitalizing on their vast pre-trained scientific priors. However, existing frameworks predominantly treat the LLM as a static generator, relying on prompt-level guidance to steer exploration. This paradigm fails to update the model's internal representations based on search feedback, often yielding physically inconsistent or mathematically redundant expressions. In this work, we propose PiT-PO (Physics-informed Token-regularized Policy Optimization), a unified framework that evolves the LLM into an adaptive generator via reinforcement learning. Central to PiT-PO is a dual-constraint mechanism that rigorously enforces hierarchical physical validity while simultaneously applying fine-grained, token-level penalties to suppress redundant structures. Consequently, PiT-PO aligns LLM to produce equations that are both scientifically consistent and structurally parsimonious. Empirically, PiT-PO achieves state-of-the-art performance on standard benchmarks and successfully discovers novel turbulence models for challenging fluid dynamics problems. We also demonstrate that PiT-PO empowers small-scale models to outperform closed-source giants, democratizing access to high-performance scientific discovery.

cs.LG↗
arXiv2026

Beyond Squared Error: Exploring Loss Design for Enhanced Training of Generative Flow Networks

Generative Flow Networks (GFlowNets) are a novel class of generative models designed to sample from unnormalized distributions and have found applications in various important tasks, attracting great research interest in their training algorithms. In general, GFlowNets are trained by fitting the forward flow to the backward flow on sampled training objects. Prior work focused on the choice of training objects, parameterizations, sampling and resampling strategies, and backward policies, aiming to enhance credit assignment, exploration, or exploitation of the training process. However, the choice of regression loss, which can highly influence the exploration and exploitation behavior of the under-training policy, has been overlooked. Due to the lack of theoretical understanding for choosing an appropriate regression loss, most existing algorithms train the flow network by minimizing the squared error of the forward and backward flows in log-space, i.e., using the quadratic regression loss. In this work, we rigorously prove that distinct regression losses correspond to specific divergence measures, enabling us to design and analyze regression losses according to the desired properties of the corresponding divergence measures. Specifically, we examine two key properties: zero-forcing and zero-avoiding, where the former promotes exploitation and higher rewards, and the latter encourages exploration and enhances diversity. Based on our theoretical framework, we propose three novel regression losses, namely, Shifted-Cosh, Linex(1/2), and Linex(1). We evaluate them across three benchmarks: hyper-grid, bit-sequence generation, and molecule generation. Our proposed losses are compatible with most existing training algorithms, and significantly improve the performances of the algorithms concerning convergence speed, sample diversity, and robustness.

cs.LG↗
arXiv2026

Charge-Transfer Induced Reactivity in sp Carbon Atomic Wires: Towards 0-D sp-sp2 Nanostructures

Carbon Atomic Wires (CAWs) are finite linear chains of sp-hybridized carbon atoms. Here the electrochemical reduction of CAWs in the form of polyynes (i.e. with alternated single-triple bonds) is reported. Upon applying a reducing potential to a solution containing polydispersed hydrogen-capped polyynes, the formation of a black precipitate was observed. Electronic absorption spectroscopy confirmed the irreversible reaction of the carbon chains while excluding degradation or side reactions. Subsequent analyses revealed that the precipitate consisted of amorphous carbon nanoparticles with tunable diameters. This control over particle size is attributed to the modulation of growth kinetics through restricted mass transport toward the solid-liquid interface. Raman spectroscopy showed that the resulting material exhibits an amorphous sp-sp2 character, with a retained sp fraction exceeding 60%. Smaller nanoparticles displayed reduced disorder within the sp2 domains and a broader distribution of sp-chain lengths preserved in the amorphous matrix. Additional experiments on size-selected polyynes suggest that this synthesis method allows to better preserve the starting chain length in the final structure. Unlike previously reported amorphous sp-sp2 carbon networks, the nanoparticles produced in this study show remarkable stability under ambient conditions, retaining their sp character for times in excess of six months. These findings pave the way for future applications, particularly as further diameter tuning may enable access to the quantum-dot regime.

cond-mat.mtrl-sci↗
arXiv2026

RefineEvo: Planning-Guided Heuristic Evolution with Bidirectional Experience

Automatic Heuristic Design (AHD) has emerged as a transformative approach for solving combinatorial optimization problems. While recent Large Language Model (LLM)-based methods have shown promise, they predominantly rely on fixed evolutionary operators and struggle to effectively accumulate and reuse historical search experience. This paper proposes RefineEvo, a novel evolutionary framework that transforms AHD from a static trial-and-error process into a planning-guided, experience-driven system. RefineEvo introduces a Planner to dynamically schedule evolutionary operators and trigger refinement based on the current search state, and a Reflector to distill valuable lessons into a Bidirectional Experience Pool containing both positive insights and negative pitfalls. This synergistic framework enables the system to adapt its search tools to the evolving complexity of the problem and leverage trajectory-aware, situation-conditioned insights to guide generation. Experiments on several classic combinatorial optimization benchmarks demonstrate that RefineEvo consistently outperforms strong baselines. In particular, RefineEvo delivers superior solution quality while improving token efficiency, enabling more efficient and autonomous heuristic design.

cs.CL↗
arXiv2026

DPO-F+: Aligning Code Repair Feedback with Developers' Preferences

Large Language Models (LLMs) are increasingly used in software engineering tasks, especially code repair. However, developers often struggle to interpret model outputs, limiting effective human--AI teaming, where humans and AI work toward a shared objective. Prior work mainly optimizes generated code, giving less attention to natural-language feedback that supports comprehension and iterative improvement. We present \textsc{DPO-f+}, a framework that aligns code-repair feedback with the needs of different developer groups, including novices and proficient developers. It (1) defines feedback-alignment metrics across seven fixed dimensions with task-specific descriptions; (2) automatically constructs pairwise preference datasets from code-repair tasks; (3) fine-tunes models using Direct Preference Optimization (DPO) augmented with a reward model; and (4) provides an automated protocol for evaluating feedback quality. Empirically, \textsc{DPO-f+} outperforms both the baseline and standard DPO in feedback accuracy and overall alignment. On novice programming tasks, \textsc{DPO-f+} improves Pass@1 by 5.71 percentage points (pp) over the baseline and 3.30 pp over DPO. On \textsc{SWE-Bench}, it improves issue-resolution rate by 1.67 pp over DPO and 4.67 pp over the baseline. It also improves feedback alignment by both LLM judges and a human study with 200 developers: beginners preferred \textsc{DPO-f+} in 71.5% of comparisons, with overall preference above chance ($p=0.0057$). By better aligning feedback with developer needs, \textsc{DPO-f+} turns LLM assistance from a one-shot output into a collaborative sense-making workflow, enhancing code comprehension and human--AI teaming in software engineering.

cs.SE↗
arXiv2026

SCOPE: Leveraging Subgoal Critiques for Code Generation

Code generation with large language models (LLMs) remains unreliable because generated programs can appear correct while still violating key semantic requirements in the natural language specification. Existing feedback-based methods improve over coder-only generation, but they often rely on unstructured critique or execution signals that do not explicitly identify what the code is semantically missing. We present SCOPE, a prover-initialized subgoal critic for code generation. SCOPE adapts a Lean-oriented prover model to produce three parseable feedback fields for downstream code generation: subgoals, gap analysis, and a robustness checklist. Our approach combines supervised fine-tuning, process-aligned reinforcement learning (RL), and feedback-guided inference, with two complementary rewards during RL: a dense reward for structured critique quality and a sparse reward based on whether the critique improves the coder's execution score. Experiments show that SCOPE improves over the compared feedback baselines. On LiveCodeBench V6, SCOPE achieves 39.4% pass@1, compared with 36.6% for Reflexion and 20.6% for the coder-only baseline. On BigCodeBench (Hard), it reaches 42.6%, surpassing Reflexion at 36.5% and coder-only generation at 34.5%. Further analysis shows that SCOPE's gains are concentrated in tasks with concrete semantic constraints and that its code corrections are more localized than Reflexion's.

cs.SE↗
arXiv2026

SearchEyes: Towards Frontier Multimodal Deep Search Intelligence via Search World Simulation

Training multimodal search agents to perform multi-hop reasoning remains challenging due to a fundamental structural disconnect: existing pipelines construct training data, search environments, and reward signals independently, causing synthesized structural metadata to be discarded, environments to rely on irreproducible external engines, and RL rewards to remain sparse at the trajectory level. We present \textbf{SearchEyes}, which uses a typed knowledge graph as the backbone of a \emph{simulated search world} that unifies all three components. We propose \textbf{Perception-Knowledge Chains (PKC)} to sample constrained multi-hop paths over the visual-knowledge intersection of Wikidata5M, retaining hop-level entity metadata that simultaneously defines a self-contained search world and step-level reward anchors. We further propose \textbf{Hop-Anchored Policy Optimization (HaPO)}, which reuses these anchors for step-level credit assignment without a separately trained process reward model. Experiments on six multimodal knowledge-intensive benchmarks show that SearchEyes achieves state-of-the-art performance among open-source multimodal search agents, with SearchEyes-27B improving over the strongest open-source baseline by 6.2 points on average.%

cs.AI↗
arXiv2026

UASPL: Uncertainty-Aware Self-Paced Learning with Evidential Neural Networks

Self-paced learning (SPL) is an effective learning paradigm that simulates the human learning process by progressing from easy to difficult samples based on the value of the loss function during the learning process. It has shown great potential in improving model performance and training efficiency. However, the prediction results of samples with smaller loss values are not necessarily reliable, indicating that such samples are not always simple samples for the model. Hence, this article proposes an uncertainty-aware self-paced learning based on evidential neural networks, termed UASPL, which integrates predictive reliability into sample selection through a general loss function within the Subjective Logic framework. This loss function incorporates uncertainty estimation and can be extended to different variants of SPL. Moreover, this loss function couples a sample selection preference, thereby ensuring the interpretability of the sample selection process. Finally, the experimental results on multiple datasets show that UASPL outperforms other SPL methods in terms of classification performance, interpretability, and generality. The source code is available at: https://github.com/treelife979/UASPL.

cs.LG↗
arXiv2026

VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Models

We present VLMEvalKit: an open-source toolkit for evaluating large multi-modality models based on PyTorch. The toolkit aims to provide a user-friendly and comprehensive framework for researchers and developers to evaluate existing multi-modality models and publish \textbf{reproducible} evaluation results. In VLMEvalKit, we implement over 450+ large multi-modality model configurations, including both proprietary APIs and open-source models, and support 330+ benchmarks across diverse multi-modal benchmarks. By implementing a single interface, new models can be easily added to the toolkit, while the toolkit automatically handles the remaining workloads, including data preparation, distributed inference, prediction post-processing, and metric calculation. VLMEvalKit has also evolved to a broader evaluation suite spanning video/audio, document understanding, GUI grounding, spatial reasoning, safety, scientific reasoning, and multi-turn dialogue. Based on the evaluation results obtained with the toolkit, we host the OpenVLM Leaderboard, a comprehensive leaderboard to track the progress of multi-modality learning research. The toolkit is released on https://github.com/open-compass/VLMEvalKit and is actively maintained.

cs.CV↗
arXiv2026

Extreme Adaptive Transformer for Time Series Forecasting

Time series forecasting remains challenging when the underlying data contain rare but critical extreme events. This issue is particularly important in hydrologic forecasting, where streamflow distributions are often highly skewed and extreme peaks can have substantial impacts on flood monitoring, water resource management, and early warning systems. Although Transformer-based forecasting models have achieved strong performance by modeling long-range temporal dependencies, they typically treat all time points uniformly and may therefore underrepresent rare extreme patterns. In this paper, we propose the Extreme-Adaptive Transformer (Exformer), a forecasting framework designed to explicitly model temporal dependencies involving both normal and extreme events. Exformer introduces an extreme-adaptive attention mechanism composed of three sparse components: Local, Stride, and Extreme. The Local and Stride components capture short-term and periodic temporal dependencies, respectively, while the Extreme component selectively models event-aware dependencies between normal and extreme streamflow patterns. Experiments on four real-world hydrologic streamflow datasets show that Exformer achieves superior 3-day forecasting performance compared with state-of-the-art baselines. Our findings demonstrate that explicitly incorporating extreme-aware attention improves the forecasting capacity of Transformer models on imbalanced time series with rare but consequential events.

cs.LG↗
arXiv2026

EyeMulator: Improving Code Language Models by Mimicking Human Visual Attention

Code Language Models (CodeLLMs) learn token importance from data correlations, whereas human developers attend selectively to semantically salient code. We present EyeMulator, a model-agnostic method that injects human visual-attention priors into CodeLLM fine-tuning without architectural changes. EyeMulator distills eye-tracking data into semantic salience and gaze-transition priors, then uses them to reweight token-level training losses. Across six backbones, two data regimes, and three CodeXGLUE tasks, the reported configurations yield positive matched-metric deltas in all 36 model-task-setting cells. Effects are largest for structure-preserving completion and translation, while summarization shows smaller but positive METEOR deltas. Session-mode and component-ablation analyses further show that reading, writing, semantic, and transition-derived priors provide complementary signal. Human-attention artifacts are available at https://zenodo.org/records/17205682.

cs.SE↗
arXiv2026

SynthFix: Adaptive Neuro-Symbolic Code Vulnerability Repair

Large Language Models (LLMs) can generate plausible code patches, but plausibility is not enough for automated repair: a patch must compile, pass tests, and remove the target vulnerability. We present SynthFix, a neuro-symbolic repair framework that combines supervised repair learning with compiler-informed feedback. During training, a lightweight router selects between Supervised Fine-Tuning (SFT) for common repair patterns and Reward Fine-Tuning (RFT) for examples that benefit from symbolic feedback. The reward combines static structure, lint/compile checks, security scanning, and public execution tests where available; at inference time, the same evidence guides best-of-K candidate selection under a greedy floor. Across five code LLMs (1.3B-7B) on pyrepair, CodeFlaws, and SVEN, SynthFix improves deployable repair metrics over SFT-only and RFT-only baselines, with relative gains up to 54 percent in functional correctness and 14 percent in security clearance. Our code and data are available at https://github.com/CoderDoge1108/SynthFix.

cs.SE↗
arXiv2026

Monadic Context Engineering

The proliferation of Large Language Models (LLMs) has catalyzed a shift towards autonomous agents capable of complex reasoning and tool use. However, current agent architectures are frequently constructed using imperative, ad hoc patterns. This results in brittle systems plagued by difficulties in state management, error handling, and concurrency. This paper introduces Monadic Context Engineering (MCE), a novel architectural paradigm leveraging the algebraic structures of Functors, Applicative Functors, and Monads to provide a formal foundation for agent design. MCE treats agent workflows as computational contexts where cross-cutting concerns, such as state propagation, short-circuiting error handling, and asynchronous execution, are managed intrinsically by the algebraic properties of the abstraction. We demonstrate how Monads enable robust sequential composition, how Applicatives provide a principled structure for parallel execution, and crucially, how Monad Transformers allow for the systematic composition of these capabilities. This layered approach enables developers to construct complex, resilient, and efficient AI agents from simple, independently verifiable components. We further extend this framework to describe Meta-Agents, which leverage MCE for generative orchestration, dynamically creating and managing sub-agent workflows through metaprogramming.

cs.AI↗
arXiv2026

HieDG: A Hierarchical Discrete Geometry-Guided Framework for Multi-Animal Tracking

Multi-animal tracking (MAT) is critical for wildlife monitoring and behavioral analysis, yet remains challenging due to uniform appearance, high density, and irregular motion. Existing methods typically follow heuristic- or query-based paradigms: the former relies on handcrafted geometric associations without end-to-end optimization, whereas the latter enables joint optimization but relies heavily on appearance embeddings. In such conditions, continuous geometric embeddings can be unstable, as small coordinate perturbations may disproportionately alter cross-frame attention weights, degrading identity association performance. To address this limitation, we propose HieDG, a Hierarchical Discrete Geometry-guided tracking framework that reformulates geometric dynamics as structured discrete representations within a query-based tracker. Instead of directly using raw geometric signals, HieDG employs a two-stage residual codebook to discretize position, scale, and velocity cues, transforming unstable continuous geometry into structured, stable discrete tokens. These tokens are aligned with visual embeddings and integrated into the tracking queries to enhance identity consistency. Extensive experiments on animal-specific benchmarks (AnimalTrack, BFT, and BuckTales) demonstrate state-of-the-art association performance with significant improvements in HOTA, AssA, and IDF1. Additional evaluations on generic multi-object tracking benchmarks, including DanceTrack and SportsMOT, show competitive performance, indicating the broader applicability of discretized geometric modeling beyond animal-specific scenarios.

cs.CV↗
arXiv2026

Causal Reward World Models: Zero-shot Reward Design for Automated Skill Generation

Automated Reward Design (ARD) aims to replace manual reward engineering in reinforcement learning with language-driven reward function synthesis. However, existing approaches based on large language models (LLMs) remain inherently correlation-driven, relying on iterative environmental feedback to refine reward hypotheses for each specific task. This paradigm not only results in inefficient reasoning but also makes LLMs susceptible to semantically plausible yet causally spurious reward components, leading to ineffective optimization. To address these limitations, we propose the Causal Reward World Model (CRWM), which explicitly models the causal topological relationships between candidate reward components and task-targeted physical variables through offline pre-training on multi-task interaction data. Based on a coarse-to-fine pre-training strategy, we introduce a joint optimization module that integrates Explicit Mechanism Decoupling with Confidence-Aware Soft Fusion to refine coarse structural priors using micro-level trajectories, thereby constructing a robust and interpretable causal skeleton. During inference, LLMs leverage CRWM as a task-irrelevant causal prior to constrain the reward generation, enabling zero-shot reward function design. Our work opens up a new white-box paradigm for the ARD problem. Extensive experiments on complex continuous control benchmarks demonstrate that CRWM generates executable reward functions without feedback-driven reward refinement, significantly reducing the design latency for acquiring new robotic skills while matching or surpassing state-of-the-art performance, and further exhibits strong generalization capabilities across unseen tasks and diverse robotic embodiments.

cs.RO↗
arXiv2026

Joint Visible Light and RF Backscatter Communications for Ambient IoT Network: Fundamentals, Applications, and Opportunities

The rapid growth of the Internet of Things (IoT) devices in the sixth generation (6G) wireless networks raises significant sustainability and scalability challenges due to energy consumption, deployment complexity, and environmental impact. Ambient IoT (A-IoT), leveraging ambient energy harvesting (EH) for batteryless device operation, has emerged as a promising solution to address these challenges. Among various EH and communication techniques, visible light communication (VLC) integrated with ambient backscatter communication (AmBC) offers remarkable advantages, including energy neutrality, high reliability, and enhanced security. In this article, we propose a joint VLC-AmBC architecture, emphasizing fundamental concepts, system designs, and practical implementations. We explore potential applications in environmental monitoring, healthcare, smart logistics, and secure communications. We present proof-of-concept demonstrations for three distinct types of ambient backscatter devices (AmBDs): EH-Only, VLC-Relay, and VLC-Control. Experimental results demonstrate the feasibility of implementing joint VLC-AmBC systems, highlighting their practical viability across various deployment scenarios. Finally, we outline future research directions, including integrated sensing and communication, as well as optimized energy-efficient deployment. Open issues, such as large-scale deployment challenges, are also discussed, thereby providing a clear roadmap for future developments in joint VLC-AmBC-enabled A-IoT ecosystems.

eess.SY↗
arXiv2026

VERITAS: Verifier-Guided Proof Search for Zero-Shot Formal Theorem Proving

LLM-based formal provers often collapse rich verifier signals (syntax errors, type mismatches, partial goal progress) into a binary pass/fail bit. We present VERITAS, a zero-shot framework that routes every verifier signal back into proof search through a two-phase protocol: Best-of-N sampling first, then a critic-guided MCTS pass that ingests Phase 1 failures as explicit negative examples. The protocol preserves every theorem solved by its own Phase 1 sweep, so Phase 2's additional solves are attributable to feedback-driven exploration. VERITAS reaches 40.6% on miniF2F (vs. an independently run Best-of-5 at 36.9%, Portfolio 26.2%) and 7.3% on VERITAS-CombiBench, a 55-theorem combinatorics benchmark we release on which Best-of-5 (1.8%) falls below Portfolio (3.6%), exposing that unguided sampling hurts when correct lemma names must be recovered iteratively from verifier feedback. Artifacts are available on GitHub.

cs.LG↗
arXiv2026

Zeta: Dual Whitening for Matrix Optimization via Coordinate-Adaptive Preconditioning

Large-scale neural network training increasingly relies on matrix-aware optimizers that exploit the structure of weight parameters beyond element-wise adaptation. However, existing matrix-aware methods such as Muon have an underappreciated vulnerability: their core operation, Newton-Schulz iteration, depends critically on input conditioning, yet the raw momentum matrices exhibit severe coordinate-wise scale heterogeneity. In this paper, we first verify this scale heterogeneity through a chi-square uniformity test, showing that intra-matrix scale imbalance is prevalent across Transformer layers and that coordinate whitening effectively corrects it. Motivated by this finding, we propose Zeta, a dual whitening optimizer that applies coordinate whitening and spectral whitening in a strictly ordered pipeline. The ordering is not a tunable choice but follows from a mathematical dependency: coordinate whitening establishes the statistical isotropy that spectral whitening requires to function reliably. We further prove that this dual pipeline strictly reduces orthogonalization error relative to pure spectral methods by improving the condition number of the input. Empirically, Zeta matches or surpasses strong baselines across language modeling (0.6B to 8B parameters), mixture-of-experts architectures, and vision tasks, demonstrating that resolving scale imbalance before orthogonalization leads to faster convergence and better generalization. Code is available at https://github.com/AIGCodeOS/aigcode_zeta_optimizer.

cs.LG↗