arXiv subjects

Wei Yang

Publications and source records attributed to Wei Yang.

At least 73 records · Page 4Linked to original sources

arXiv2026

Correct but Slow: An Empirical Study of the GPU Kernel Evaluation Gap in Modern Domain-Specific Languages

Modern GPU domain-specific languages (DSLs), such as Triton and TileLang, are increasingly used to implement specialized deep-learning kernels and as target languages for automated kernel-generation systems. Existing DSL-kernel evaluations establish correctness through reference-based numerical validation -- necessary, but silent on replacement quality: a functionally valid kernel may still fall far below the throughput of the optimized library operator it is intended to replace. We study this correctness-performance gap using 22 Triton and TileLang kernels from five operator categories on NVIDIA A100 and GH200 GPUs, asking whether correctness-based evaluation identifies kernels unsuitable as library replacements, why such failures occur, and how they can be detected without exhaustive benchmark coverage. The study yields three results. \emph{First}, correctness-based evaluation can admit severe slowdowns: an idiomatic TileLang LayerNorm kernel passes KernelBench's correctness check while running more than 300$\times$ slower than the PyTorch baseline. \emph{Second}, the causes differ by kernel family. TileLang normalization and reduction slowdowns are mainly repairable authoring defects, such as sequential reductions and unnecessary dtype conversions, whereas convolution and large general matrix multiplication (GEMM) retain residual gaps after optimization due to code-generation and autotuning-coverage limits; vendor-library algorithm selection contributes only marginally. \emph{Third}, two lightweight checks -- library-relative efficiency and roofline utilization -- are complementary screening criteria: together they flag every functionally valid but inefficient kernel in our suite and separate repairable authoring defects from structural residuals.

cs.SE↗

arXiv2026

BackendForge: Benchmarking Agentic End-to-End Code Generation with Backend Services

Large language models (LLMs) are increasingly used in agentic coding settings, where they can inspect files, execute commands, run tests, observe failures, and iteratively revise code. This shift raises a central evaluation question: can an agentic LLM generate an end-to-end software artifact that is both deployable and behaviorally correct under execution? Backend services provide a controlled but realistic substrate for this evaluation. Their APIs expose application-level executable semantics, and deployed behavior can be checked deterministically against an OpenAPI contract through black-box HTTP interactions. We introduce BackendForge, a benchmark of 56 contract-defined backend generation tasks rewritten from real open-source applications. Given a visible specification and an OpenAPI contract, an LLM must generate a Dockerized service that is built, deployed, and evaluated only through HTTP tests. To strengthen evaluation without introducing hidden requirements, BackendForge uses a test agent and a code agent to co-evolve the test oracle and reference service, where the test agent proposes specification-grounded backend tests and the code agent repairs the reference implementation. Although the best-performing model, GPT-5.5, succeeds on 55.4\% of tasks under the base oracle, it succeeds on only 28.6\% under the final oracle. This gap suggests that current LLMs can implement many local API behaviors, but still struggle to produce complete backend services.

cs.SE↗

arXiv2026

ViPSim: Collaborating Visual and Parameter Spaces for Consistent Long-Horizon Embodied World Models

Embodied World Models (EWMs) have emerged as a scalable and risk-free paradigm for advancing embodied intelligence, enabling the safety-critical evaluation of Vision-Language-Action systems. However, their reliability as evaluation benchmarks and foundational simulators is often hindered by the representation gap between low-dimensional actions and high-dimensional video synthesis. This gap results in a lack of geometric correspondence, manifesting as accumulated trajectory drift and inconsistent robot-object interactions during long-horizon rollouts. To bridge this gap, we propose ViPSim, a framework that achieves consistent long-horizon generation through the synergistic collaboration of Visual and Parameter Spaces. We define the Visual Space as a domain of explicit spatial priors, integrating pixel-aligned projections of end-effector pose, camera perspectives, depth-informed scene geometry, and robotic morphological masks to provide dense structural grounding. Concurrently, the Parameter Space serves as a domain of numerical drivers, injecting raw action sequences and camera matrices to provide precise motion guidance. By unifying these two spaces, ViPSim ensures that the generated states are simultaneously anchored by geometric boundaries and steered by numerical commands. Extensive experiments demonstrate that ViPSim is backbone-agnostic and significantly enhances trajectory consistency. Notably, our approach exhibits emergent capabilities in generating complex interactions with deformable objects (e.g., cloth folding) and maintains robust performance in out-of-distribution and cross-embodiment scenarios, providing a high-fidelity foundation for the automated evaluation and predictive control of embodied agents.

cs.CV↗

arXiv2026

ReM-MoA: Reasoning Memory Sustains Mixture-of-Agents Scaling

Mixture-of-Agents (MoA) architectures improve inference-time scaling by organizing multiple LLM agents into layered reasoning pipelines. However, existing MoA variants fail to sustain gains as depth increases, exhibiting degradation, early plateauing, or saturation. We propose ReM-MoA, a memory-augmented MoA framework that sustains scaling through two mechanisms: (1) a Ranked Reasoning Memory that persistently stores and ranks reasoning traces from all layers using a comparative Reviewer Agent, and (2) a Curated Diversified Memory Routing scheme that exposes different agents to distinct combinations of successful and failed traces, preserving exploration diversity while propagating high-quality reasoning. We further introduce an optional multi-domain Reviewer distillation pipeline that improves ranking quality through frontier-model supervision. Across five reasoning benchmarks spanning math, formal logic, code, knowledge, and commonsense, ReM-MoA consistently outperforms prior MoA variants across both depth and width scaling, and its advantage widens with depth, establishing structured cross-layer reasoning memory as a key missing mechanism for scalable multi-agent inference.

cs.AI↗

arXiv2026

RaMem: Contextual Reinstatement for Long-term Agentic Memory

Long-term memory has become increasingly important for LLM agents that operate across extended interactions and evolving task contexts. Recent memory systems have made past experiences more persistent, compact, and retrievable, but retrieval alone does not ensure that a memory provides valid evidence for the current query. When experiences are compressed into reusable fragments, memories from different situations may appear equally relevant if they involve recurring entities or user states. We refer to this failure as context collapse: memories lose the surrounding context needed to judge whether they provide valid evidence for the current query. To address this problem, we propose Contextual Reinstatement for Agentic Memory (RaMem), a framework that turns retrieved memory fragments into contextually verifiable evidence. RaMem operates through four coordinated stages: (i) evidence anchoring grounds each memory in its original episodic conditions, especially event time, mention time, session span, and participants; (ii) recall condition induction derives the evidence conditions implied by the query; (iii) validity-aware retrieval uses these conditions to prioritize context-compatible memories while retaining content-relevant candidates as fallback evidence; and (iv) context-preserved synthesis keeps the selected memories' structured context available to the generator. Experiments on long-term memory benchmarks show that RaMem consistently improves performance over strong memory baselines, with average F1 gains of more than 10% across several backbones.

cs.AI↗

arXiv2026

Understanding Automated Web GUI Testing: An Empirical Study Across Exploration Strategies and State Abstractions

Automated web GUI testing (AWGT) relies on exploration strategies that exercise web applications through GUI actions to maximize code coverage, spanning traditional model-based, reinforcement learning (RL)-based, and emerging large language model (LLM)-based approaches. State abstraction, which detects pages with the same functionality to avoid repeated testing, has long been recognized as critical to guiding exploration. However, how exploration strategies and state abstractions jointly affect testing effectiveness remains underexplored. We present an empirical study analyzing both factors from the perspectives of code coverage and failure revelation. We compare representative model-based, RL-based, and LLM-based approaches; investigate how six state abstractions influence model-based and RL-based approaches; examine LLM-based approaches under different history representations, which act as a form of state abstraction; and compare the failures exposed by different approaches. Our results show that no single strategy excels across all dimensions; instead, categories exhibit complementary strengths in code coverage, state coverage, and failure discovery. State abstraction is a key factor: strict, fine-grained abstractions favor model-based strategies, while compact ones better support RL-based strategies. History representation substantially affects LLM-based strategies, where concise, functionality-level context performs best. We also find that code coverage is weakly correlated with failure-revealing ability, underscoring the need for multi-dimensional evaluation. These findings offer practical guidance for selecting exploration strategies and designing effective state abstractions for AWGT.

cs.SE↗

arXiv2026

FORTIS: Benchmarking Over-Privilege in Agent Skills

Large language model agents increasingly operate through an intermediate skill layer that mediates between user intent and concrete task execution. This layer is widely treated as an organizational abstraction, but we argue it is also a privilege boundary that current models routinely exceed. We present \textbf{FORTIS}, a benchmark that evaluates over-privilege in agent skills across two stages: whether a model selects the minimally sufficient skill from a large overlapping library, and whether it executes that skill without expanding into broader tools or actions than the skill permits. Across ten frontier models and three domains, we find that over-privileged behavior is the norm rather than the exception. Models consistently reach for higher-privilege skills and tools than the task requires, failing at both stages at rates that remain high even for the strongest available models. Failure is especially severe under the ordinary conditions of real user interaction: incomplete specification, convenience framing, and proximity to skill boundaries. None of these requires adversarial construction. The results indicate that the skill layer, far from containing agent behavior, is itself a primary source of privilege escalation in current systems.

cs.AI↗

arXiv2026

A Survey of Reasoning and Agentic Systems in Time Series with Large Language Models

Time series reasoning treats time as a first-class axis and incorporates intermediate evidence directly into the answer. This survey defines the problem and organizes the literature by reasoning topology with three families: direct reasoning in one step, linear chain reasoning with explicit intermediates, and branch-structured reasoning that explores, revises, and aggregates. The topology is crossed with the main objectives of the field, including traditional time series analysis, explanation and understanding, causal inference and decision making, and time series generation, while a compact tag set spans these axes and captures decomposition and verification, ensembling, tool use, knowledge access, multimodality, agent loops, and LLM alignment regimes. Methods and systems are reviewed across domains, showing what each topology enables and where it breaks down in faithfulness or robustness, along with curated datasets, benchmarks, and resources that support study and deployment (https://github.com/blacksnail789521/Time-Series-Reasoning-Survey). Evaluation practices that keep evidence visible and temporally aligned are highlighted, and guidance is distilled on matching topology to uncertainty, grounding with observable artifacts, planning for shift and streaming, and treating cost and latency as design budgets. We emphasize that reasoning structures must balance capacity for grounding and self-correction against computational cost and reproducibility, while future progress will likely depend on benchmarks that tie reasoning quality to utility and on closed-loop testbeds that trade off cost and risk under shift-aware, streaming, and long-horizon settings. Taken together, these directions mark a shift from narrow accuracy toward reliability at scale, enabling systems that not only analyze but also understand, explain, and act on dynamic worlds with traceable evidence and credible outcomes.

cs.AI↗

arXiv2026

Supermoiré Chern mosaic in helical trilayer WSe2

Helically twisted multilayers offer access to moiré physics beyond the single-superlattice paradigm, yet their correlated and topological transport properties remain largely unexplored in semiconductor moiré materials. Here we report magnetotransport measurements of helical trilayer WSe2, in which two coupled moiré patterns relax into a supermoiré landscape composed of inequivalent local topological domains with distinct electronic structures and unequal spatial areas. By electrostatic tuning, we identify a trilayer-hybridized regime where interactions and real-space reconstruction combine to generate a plethora of magnetic and topological states absent in the twisted bilayers. At moiré filling factor $ν$ = -1, we observe a ferromagnetic insulating state that is robust against magnetic field and accompanied by a non-quantized anomalous Hall response ~-4 kOhms. This behaviour is consistent with a time-reversal-symmetry-breaking supermoiré Chern mosaic, in which the Hall response arises from the non-cancelling contributions of local domains with opposite Chern character arranged by the relaxed structure. Under strong magnetic fields, a symmetry-broken Chern insulating state (C = 1) emerges near $ν$ = -2/3, displaying a much larger positive Hall response together with strongly enhanced longitudinal resistance, suggestive of field-reconstructed topological minibands and domain-boundary scattering. These results establish relaxed supermoiré semiconductor trilayers as a platform for spatially organized magnetism and topology beyond the bilayer limit.

cond-mat.str-el↗

arXiv2026

Food-R1: A Unified Multi-Task Food Vision-Language Model with Reinforcement Learning

Recent studies have explored Vision-Language Models (VLMs) for food analysis. However, most existing methods rely primarily on supervised fine-tuning (SFT), which often limits reasoning and generalization capabilities. Moreover, high-quality large-scale nutritional annotations remain scarce. To address these issues, we introduce CalorieBench-80K, a large-scale benchmark with curated calorie labels and dietary advice annotations. To the best of our knowledge, it is the first food image benchmark to incorporate Chain-of-Thought (CoT) annotations for calorie reasoning. We also propose Food-R1, a unified food VLM trained in a multi-task learning paradigm to equip the model with broad capabilities. Food-R1 undergoes CoT-based cold-start instruction tuning, followed by reinforcement fine-tuning (RFT) using Group Relative Policy Optimization (GRPO) to improve reasoning and performance. Experiments on CalorieBench-80K and representative benchmarks show that Food-R1 consistently outperforms strong baselines across food-related tasks. The code, model weights, and benchmark annotations are available at the project repository.

cs.CV↗

arXiv2026

Memory Retrieval for Changing Preferences

Long-context dialogue systems must decide both when to access memory and which parts of the interaction history are relevant. Existing approaches typically rely on heuristic retrieval signals or always-on memory usage, failing to account for the changing and potentially inconsistent nature of user preferences. In this work, we propose a unified framework for memory access and selection based on changing preferences. We formulate personalized memory retrieval as identifying which historical turns provide evidence about a user's latent preference state, rather than relying on surface-level semantic similarity. To this end, we quantify the utility of each memory turn using a Bayes factor, defined as the improvement in the model's likelihood of the reference response when the turn is included in context. This provides a principled measure of evidence strength and a unified signal for both memory access and selection. By framing memory retrieval as utility estimation, the model learns to identify salient turns and regulate memory usage based on expected utility. Experiments on four heterogeneous memory benchmarks show that our approach outperforms existing embedding-based retrieval on long-context, preference-intensive tasks where modeling changing preferences is essential, while remaining competitive in low-density regimes where semantic similarity suffices.

cs.CL↗

arXiv2026

PRISM: Synergizing Vision Foundation Models via Self-organized Expert Specialization

Unifying the complementary strengths of diverse Vision Foundation Models (VFMs) into a single efficient model is highly desirable but challenged by the negative transfer inherent in monolithic distillation. To address these feature conflicts, we introduce \textbf{PRISM}, a novel dual-stream Mixture-of-Experts (MoE) framework that synergizes VFMs via modular specialization. We propose a two-stage paradigm: (1) expertise deconstruction, where a teacher-conditional router guides experts to specialize in distinct representational subspaces to mitigate interference, followed by (2) dynamic recomposition, where the router learns to assemble these experts into tailored computational pathways for downstream tasks. Experiments on PASCAL-Context and NYUD-v2 show that \textbf{PRISM} establishes a new state of the art, validating that sparse, emergent specialization is a scalable approach for integrating diverse visual knowledge.

cs.CV↗

arXiv2026

Ferroelectric-tunable quantum nonlinearity of chiral Bloch electrons in a moiré system

Sliding ferroelectricity in van der Waals materials shows great potential for designing robust memory devices. However, its thermodynamic behaviors and the coupling with certain quantum effects remain largely unexplored. Here, we demonstrate ferroelectric control over quantum nonlinear transport in a hexagonal boron nitride (hBN) encapsulated twisted double-bilayer graphene moiré heterostructure. The ferroelectricity is attributed to the presence of rhombohedral stacking in the top hBN, confirmed by both electrical transport and optical second harmonic generation (SHG) measurements. Remarkably, the polarization magnitude remains temperature-independent across 1.7-200 K, while nucleation time exhibits thermally activated behavior, decreasing with increasing temperature. Furthermore, we demonstrate a ferroelectric-switchable nonlinear Hall effect, attributed to the chiral scattering induced by Berry curvature, with outstanding fatigue-resistant and nonvolatility, demonstrating direct coupling between sliding ferroelectricity and quantum geometric properties. Our results establish sliding ferroelectrics as a platform for exploring electrically programmable Berry curvature physics.

cond-mat.mes-hall↗

arXiv2026

UnityMAS-O: A General RL Optimization Framework for LLM-Based Multi-Agent Systems

LLM-based multi-agent systems decompose complex tasks into interacting roles, but most remain manually orchestrated by prompts, tools, and control rules, while agents are rarely optimized through a unified reinforcement learning interface. Existing RL post-training frameworks mainly target single-policy optimization and lack abstractions for user-defined multi-agent workflows, structured interaction, role-specific credit assignment, and configurable parameter sharing. We present UnityMAS-O, a general RL optimization framework for LLM-based multi-agent systems. UnityMAS-O treats the complete workflow as the optimization unit, rather than a single response or policy trajectory. It represents workflows through four first-class objects: logical agent roles, graph trajectories, user-defined rewards, and agent--model mappings. This decouples logical agents from physical model parameters, supporting full sharing, full separation, and partial sharing, with rewards assigned at role, turn, and trajectory levels. UnityMAS-O extends verl with a Ray-based star-topology runtime. A central controller executes workflows, invokes tools, records structured trajectories, and assembles rewards; model-local worker groups handle rollout, buffering, advantage computation, and distributed PPO-style updates. Users can define agents, workflows, model mappings, and rewards without rewriting the optimization infrastructure. We instantiate UnityMAS-O on retrieval-augmented QA, iterative agentic search, and reflective code generation. Across Natural Questions, HotpotQA, and held-out code tasks, multi-agent RL improves manually specified workflows after optimization, with especially large gains for smaller models and strict code all-passed metrics. These results show that UnityMAS-O can serve as a reusable substrate for converting diverse LLM-based multi-agent workflows into trainable multi-agent RL systems.

cs.AI↗

arXiv2026

RecGOAT: Graph Optimal Adaptive Transport for LLM-Enhanced Multimodal Recommendation with Dual Semantic Alignment

Integrating large language model (LLM) representations into multimodal recommendation has shown promise, yet a fundamental challenge remains largely overlooked: the semantic heterogeneity between generative LM representations and the ID-based collaborative signals that recommendation systems rely on. Naively injecting LM features without alignment degrades recommendation performance rather than improving it. To resolve this, we propose RecGOAT, a dual-granularity semantic alignment framework built on graph neural networks and optimal transport theory. RecGOAT first enriches collaborative semantics through multimodal attentive graphs that capture item-item, user-item, and user-user relationships, initializing user representations via LLM-inferred behavioral preferences. It then aligns LM-derived modality representations with recommendation IDs at two complementary granularities: (1) instance-level alignment via cross-modal contrastive learning (CMCL), which produces discriminative per-sample representations; and (2) distribution-level alignment via optimal adaptive transport (OAT), which minimizes the 1-Wasserstein distance between ID distributions and LLM semantics to produce a unified, consistently aligned feature space. Theoretically, we prove that the unified representation achieves strictly lower target error than any single-modality representation, with the gap bounded by the Wasserstein distance and the InfoNCE loss, providing rigorous guarantees for both alignment consistency and fusion comprehensiveness. Extensive experiments on three public benchmarks demonstrate state-of-the-art performance. Deployment on a large-scale online advertising platform further validates RecGOAT's industrial scalability. Our code is available at https://github.com/6lyc/RecGOAT-LLM4Rec.

cs.IR↗

arXiv2026

OASES: Outcome-Aligned Search-Evaluation Co-Training for Agentic Search

Agentic search enables language models to solve knowledge-intensive tasks by adaptively acquiring external evidence over multiple steps. Reinforcement learning with verifiable rewards (RLVR) has emerged as a widely adopted training paradigm for search agents, yet outcome-only rewards are sparse and provide limited credit assignment for intermediate search actions. Existing process-reward methods therefore seek to densify supervision through proxy signals, external evaluators, or likelihood-based information gain. However, proxy rewards can deviate from the final outcome objective, while fixed evaluators can become stale as the search policy evolves, leading to unreliable process supervision. To address these challenges, we propose OASES, an Outcome-Aligned Search-Evaluation Supervision framework for agentic search. OASES derives outcome-aligned process rewards by evaluating how well each intermediate search state supports answering the original question. It further co-trains the search policy and the state evaluator on policy, allowing the evaluator to adapt to evolving search behavior and provide more reliable process rewards. Experiments on five multi-hop QA benchmarks show that OASES consistently outperforms strong RL baselines, with further analyses confirming the benefits of outcome-aligned process rewards and search-evaluation co-training.

cs.AI↗

arXiv2026

Momentum-Resolved Tunneling Modulation Induced Giant Multistate Resistance in Antiferroelectric Multiferroic Junction

Multiferroic tunnel junctions (MFTJs), integrating ferroelectric and ferromagnetic functionalities within a single nanoscale device, hold significant promise for non-volatile, multi-state memory and innovative computing paradigms. In conventional MFTJs, tunneling resistance modulation relies primarily on ferroelectric (FE) polarization switching, which alters interfacial electric fields and shifts the Fermi level of adjacent ferromagnetic electrodes. However, achieving high tunnelelectroresistance (TER) through this approach demands strong built-in electric fields, which simultaneously hinder FE polarization switching, creating an intrinsic trade-off between reliable data reading and efficient writing. Here, we propose a dual mechanism that combines antiferroelectric (AFE) phase-transition modulation of the evanescent decay states with interfacial spin filtering based on $Fe_3GaTe_2$/bilayer-$In_2Se_3$/$Fe_3GaTe_2$ heterostructure. Beyond altering the electrostatic potential as in AFE-FE switching, the transitions between head-type and tail-type AFE states preserve the centrosymmetric potential profile yet fundamentally modulate the momentum-resolved distribution of evanescent decay rates across the Brillouin zone. When integrated with perfect spin filtering at the $Fe_3GaTe_2$/$α$-$In_2Se_3$ interface, this mechanism yields a giant TER (~$7.6\times10^3\%$), over 4 times that of conventional FE-based MFTJs, and a TMR exceeding $6.8\times10^5\%$, enhanced by two orders of magnitude over typical MFTJs. These mechanisms resolve the performance trade-off in MFTJs, enabling six distinct non-volatile resistance states at room temperature.

cond-mat.mes-hall↗

arXiv2026

On-the-Fly Input Adaptation for Reliable Code Intelligence

Code language models (CLMs) play a central role in software engineering across both generation and classification tasks. However, these models still exhibit notable mispredictions in real-world applications, even when trained on up-to-date data. Existing solutions address this by retraining the model, modifying its architecture, or re-engineering prompts. These approaches incur high computational cost requiring substantial effort in data labeling, model updates, and redeployment, and often suffer from poor generalization across tasks and tuning instability across models. This work proposes an alternative strategy based on on-the-fly input adaptation, which improves model behavior without altering its parameters or requiring additional supervision. The method consists of two stages: input validation, which detects inputs likely to cause mispredictions, and input adaptation, which transforms them using syntax- and semantics-preserving operations to better align with the model's learned behavior. This dual strategy reduces mispredictions across diverse code understanding tasks, boosting model performance without necessitating retraining. As a scalable and resource-efficient solution, this framework holds significant promise for high-stakes applications in software engineering where reliability is critical.

cs.SE↗