Search arXiv⌕ Search

arXiv subjects

Zhao Yang

Publications and source records attributed to Zhao Yang.

At least 19 recordsLinked to original sources
arXiv2026

Unstable Stokes waves in constant vorticity flows of finite depth

We study the spectral instability of small-amplitude Stokes waves in finite-depth flows with constant vorticity. A center-manifold reduction and the periodic Evans function give two explicit indices, one for long-wave instability near the origin and the other for an order-two resonance away from the origin. We prove instability when the low-frequency index is positive or the high-frequency index is nonzero, and obtain the leading expansions of the unstable spectral points. At low frequency, first- and second-order monodromy coefficients determine the leading figure-eight curve. At high frequency, we prove that the resonance exists uniquely and derive a limiting ellipse. An algebraic identity expresses the high-frequency index as a positive factor times the square of a real coupling coefficient. Thus every parameter pair with nonzero coupling gives high-frequency instability. The explicit formulas also permit a comparison of the two criteria in the parameter plane.

math.AP↗
arXiv2026

Unstable Stokes waves in constant vorticity flows of infinite depth

We study the spectral stability of small-amplitude Stokes waves with constant vorticity in infinite depth. We describe the spectrum near the origin and near the first nonzero double eigenvalue. For nonzero vorticity $ω<2\sqrt{g/3}$, where $g$ is the gravitational constant, the eigenvalues near the origin form a figure eight, whereas for $ω>2\sqrt{g/3}$ the local spectrum remains purely imaginary. Away from the origin, the interaction of two Fourier modes whose indices differ by two produces an unstable isola when $ω>-\sqrt{g/6}$, $ω\ne0$, and the resonant wave numbers are noninteger. This curve is a real analytic oval with an explicit leading ellipse and maximal growth of quadratic order in the amplitude. We also prove instability at the integer resonances needed to cover the remaining vorticities. Together with the classical irrotational result, these two mechanisms show that sufficiently small nonzero Stokes waves are spectrally unstable for every fixed constant vorticity. The proof uses finite-dimensional spectral reductions of the full Euler linearization. Near the origin, the finite-dimensional linearized operator has a one-sided limit as the Floquet exponent vanishes, which differs from the periodic linearization. We use the Kato transformation to construct a three-dimensional reduction with biorthogonal bases and use all three eigenvectors of the reference operator to calculate the Benjamin--Feir coefficient. At the nonzero collision, a real two-dimensional reduction and an explicit second-order calculation determine the unstable curve.

math.AP↗
arXiv2026

GRID-Planner: Graph-Structured Relational Reward Shaping for Diffusion Planning

Autonomous driving planning requires selecting trajectories that remain safe, rule-compliant, and comfortable under complex multi-agent interactions, yet robust closed-loop decision making in planning-critical scenes remains challenging. Existing diffusion- and RL-based planners often rely on generic rewards and use relational structure mainly for representation, limiting how planning-critical rollouts shape policy improvement. We present GRID-Planner, a graph--reward integrated diffusion planning framework that aligns generative planning with structured relational signals. Specifically, GRID-Planner combines graph-aware importance weighting, a time-indexed relational reward surrogate, and an advantage-weighted diffusion objective with on-policy self-distillation, forming a rollout-return-advantage-update loop without explicit likelihoods. Experiments on nuPlan and interPlan show that, compared with the diffusion-only baseline, GRID-Planner reduces collision rate, lane deviation, and jerk by 35.8\%, 60.8\%, and 27.4\%, respectively, and improves the planning score by 3.4\%. Project website: https://github.com/yangzhaojason/GRID-Planner.

cs.RO↗
arXiv2026

LongCat-DeepResearch Technical Report

We present LongCat-DeepResearch, a deep research system that combines an enhanced LongCat model with a multi-agent workflow for producing comprehensive, evidence-grounded reports. The workflow separates global planning from detailed investigation and coordinates revision at the section level. Multiple planning agents first explore external sources and refine an actionable research plan, termed ResearchSpec. Research agents then investigate and draft their assigned sections in parallel, gathering additional evidence in separate contexts as their analyses develop. Once the sections are assembled, global review guides targeted local revisions, reducing reliance on repeated full-report rewriting. This workflow also supports the construction of research tasks and trajectories for the mid-training and post-training of LongCat's general-purpose models. LongCat-DeepResearch achieves 55.25 on DeepResearchBench, 51.35 on DeepResearchBench II, and 79.83 on ResearchRubrics. On an in-house benchmark, it scores 76.04, ranking second among four compared systems. Development-set analyses show benefits from combining planning perspectives, while further planning refinement has mixed effects. Additional editing improves average automatic readability preference across two benchmarks, with different trends on each.

cs.AI↗
arXiv2026

ELEMENT: Episodic and Lifelong Exploration via Maximum Entropy

Reinforcement learning agents depend on reward signals whose density is rarely under the designer's control, and when such signals are absent, an agent must generate its own drive to explore. State entropy maximization offers a principled objective for this, but existing methods break down at scale in two ways: the intrinsic reward vanishes once a state has been visited, discouraging revisits to the very gateways that lead onward, and estimating entropy over millions of accumulated observations becomes computationally prohibitive. We address both with Episodic and Lifelong Exploration via Maximum Entropy (ELEMENT), a multiscale intrinsically motivated framework for reward-free exploration that transfers to downstream tasks. ELEMENT couples lifelong entropy maximization with a complementary episodic term acting on a faster timescale. For the episodic term, we derive average episodic state entropy, an intrinsic reward that is the exact minimizer of a tractable upper bound on the reward-decomposition objective; for the lifelong term, we propose a $k$NN graph-based estimator that keeps entropy tractable without forgetting. ELEMENT consistently outperforms state-of-the-art intrinsic reward baselines on state coverage and unsupervised pre-training. Videos, code, and supplementary material: https://sites.google.com/view/element-rl.

cs.LG↗
arXiv2026

Stable and unstable capillary-gravity waves

We develop a rigorous theory of spectral stability and instability for traveling periodic capillary-gravity waves of sufficiently small amplitude in a fluid of finite depth, encompassing unimodal waves and bimodal Wilton ripples. Our analysis builds upon a periodic Evans function framework developed by the authors for Stokes' gravity waves, but requires substantial extensions and new analytical tools to accommodate the considerably richer dynamics of the problem. Particularly, we show that the spectrum away from resonances remains confined to the imaginary axis for sufficiently small amplitude. A key new ingredient in the analysis for unimodal waves near nonzero resonances is an application of the Weierstrass preparation theorem to the discriminant of the Weierstrass polynomial associated with the periodic Evans function. This yields an explicitly computable real-analytic function of the amplitude, whose sign determines spectral stability or instability. Particularly, we obtain the first rigorous spectral stability and instability criteria for traveling periodic capillary-gravity waves away from the origin of the complex plane. We further show that, to leading order in the amplitude, the unstable spectrum near a nonzero resonance traces either an ellipse or a circle. We then investigate modulational instability near the origin of the complex plane for unimodal waves. Our stability index agrees with that obtained from formal asymptotic analysis and recovers the Benjamin--Feir instability for Stokes waves in the zero surface-tension limit. We extend the analysis to bimodal Wilton ripples and obtain the first rigorous spectral stability and instability criteria away from the origin. We numerically evaluate the resulting stability indices and identify parameter regimes for spectral instability either near the origin or near a nonzero resonance of order one or two.

math.AP↗
arXiv2026

Ignorance or Incompetence? Constructing Knowledge-Gated, Verifiable Tasks for LLM Agents

Professional agent tasks often depend on conventions that are absent from public corpora, yet benchmarks rarely control whether an agent has access to those conventions. We introduce a knowledge-gated task-construction protocol that separates a task instruction from a compact artefact containing private conventions, reference tables, and utility operators. Construction-time provenance, byte-identical task instructions across the provided- and withheld-artefact conditions, leak audits, and executable witnesses make dependence on the artefact explicit and testable. Across fifteen calibration tasks, one frontier agent configuration achieves a 68.0% pass rate with the artefact and 0% without it; on one task, a plausible but incorrect artefact also yields 0% across five trials. Deterministic solvers and rule corpora provide exact ground truth for structured tasks, while named criterion-level rubrics support outputs that cannot be checked by a single executable oracle. A configuration-relative calibration screen retains seven tasks satisfying our five-trial empirical knowledge-gating screen. These experiments validate the behavior of the construction protocol; they do not establish that the retained tasks improve post-training. We publicly release part of the task suite and supporting tooling at https://github.com/DatagridsAI/Knowledge-Gated-Task-Construction.

cs.AI↗
arXiv2026

Learning When to Think: Adaptive Reasoning for Test-Time Compute Allocation

Reasoning language models trained with reinforcement learning typically operate under a fixed token budget rather than an explicitly adaptive one, which can lead to over-computation on easy problems and insufficient computation on difficult ones. We study whether a model can learn to allocate its own reasoning effort by choosing, as the first token of its response, one of three modes: \textsc{NoThink} (answer as quickly as possible), \textsc{Short} (brief reasoning), or \textsc{Long} (extended reasoning). The choice is learned inside Group Relative Policy Optimization (GRPO) with no separate router, through a shaped reward that makes each mode worthwhile at a different response length, together with hard per-mode token caps that keep the modes distinct. On a 1.5B distilled model trained on MATH, the three modes emerge without collapsing to a single choice, and the brief modes end up more accurate than \textsc{Long}, which shows that the router sorts problems by difficulty rather than at random. Averaged over three seeds, the resulting policy stays close to the base model's accuracy on the held-out MATH500 ($0.782$ vs.\ $0.796$) while cutting the mean response length from $4{,}796$ to $2{,}811$ tokens (a $41\%$ reduction). Interestingly, it also transfers to other benchmarks without retraining, with the largest savings where problems are easier, with for instance 76\% token reduction on GSM8K and at higher accuracy than the baselines at similar response length. In short, we build a reasoning model that adaptively chooses how much to reason for each problem.

cs.AI↗
arXiv2026

Hermite Curves as Trajectory Priors for Vision-Language-Action Models

Despite recent progress in Vision-Language-Action (VLA) models for robotic manipulation, the action chunk remains a weakly structured interface. Existing work typically flatten each chunk into per-timestep controls, relying on implicit data learning that manifests as jagged motion and boundary discontinuities during physical execution. To address these limitations, we introduce Hermite trajectory priors, parameterizing the chunk trajectory as a piecewise cubic Hermite curve defined by endpoint positions and velocities to explicitly enforce smoothness and continuity. We instantiate this fixed operator across discrete autoregressive and continuous generative paradigms via three variants: (1) Hermite Tokens, which predict quantized boundary variables autoregressively; (2) Hermite Scaffold, which decomposes clean actions into a base scaffold and residuals; and (3) Hermite Regularization, which applies the prior strictly as an auxiliary training objective. Across simulation benchmarks and real-robot platforms, Hermite Regularization achieves superior performance among these three variants, improving π0.5 baseline success rates from 95.9% to 98.7% on LIBERO, 85.7% to 90.9% on LIBERO-plus, and 63.4% to 90.0% across four real-robot tasks without additional inference overhead. Trajectory analyses reveal that explicitly structuring trajectory priors serves most effectively as a learning inductive bias rather than a runtime constraint.

cs.RO↗
arXiv2026

SkillReason: Reasoning-Enhanced Agent Skill Retrieval for Implicit User Requests

Large language model agents increasingly rely on reusable skills to extend their capabilities beyond parametric knowl- edge. However, retrieving the appropriate skill from a large- scale library remains challenging because realistic user re- quests are often concise and underspecified, stating only the task goal while leaving the required capabilities and execu- tion steps implicit. Existing benchmarks provide limited cov- erage of such requests. To address this gap, we introduce SkillReason-Bench, a large-scale cross-domain benchmark containing 3,729 queries and a retrieval corpus of 61,228 skills spanning nine domains. We further propose SkillRea- son, a two-stage framework that uses chain-of-thought rea- soning as training-time supervision for skill retrieval. In Stage I, capability reasoning traces generated by a stronger teacher provide explicit supervision through contrastive learning, re- trieval distribution alignment, and language modeling, en- couraging the retriever to internalize capability reasoning in its query representation. In Stage II, a retrieval-guided GRPO objective encourages the model to explore reasoning trajecto- ries better suited to its own capabilities and more effective for retrieval. At inference, SkillReason directly encodes the orig- inal query without autoregressive CoT generation, preserv- ing efficient query-only retrieval. Extensive experiments on SkillReason-Bench, SkillRet, and SRA-Bench show that Skill- Reason achieves state-of-the-art performance across all three benchmarks, demonstrating that reasoning-enhanced training better bridges the semantic gap between high-level task goals and skill capabilities.

cs.AI↗
arXiv2026

Agentic Reinforcement Learning with Observation-Calibrated Self-Distillation

Large language model agents are commonly trained through reinforcement learning with sparse trajectory-level rewards, which offer limited guidance on how strongly individual tokens should be updated. On-Policy Self-Distillation (OPSD) addresses this by re-scoring generated tokens under a privileged replay view to obtain dense, token-level supervision. However, we identify a confounding issue: the resulting support may reflect both the privileged information contained in the replay view and score shifts induced by the replay scaffold, making it difficult to attribute the support specifically to that information. This issue is especially pronounced when future environment observations serve as privileged information, since replaying them requires reconstructing an extended scaffold that itself perturbs token scores. To resolve this confounding, we propose Observation-Calibrated Self-Distillation (OCSD), which contrasts two structurally matched replay views, Full and Observation-Ablated, differing only in whether the actual future observation is present, to derive an observation residual that discounts score changes shared by the replay scaffold. OCSD then applies this residual to modulate token-level GRPO updates at high-uncertainty steps, while preserving the trajectory-level update direction. Experiments on ALFWorld, WebShop, and Search-QA across three Qwen3 model scales show that OCSD consistently outperforms strong baselines. Diagnostic analyses further confirm that the calibrated residual aligns better with local environment feedback. Our code is publicly available at https://github.com/yiy1x/OCSD.

cs.LG↗
arXiv2026

ChunkFlow: Towards Continuity-Consistent Chunked Policy Learning

Vision-language action (VLA) models increasingly adopt chunked action heads to satisfy real-time constraints; however, this introduces boundary jitter: overlapping regions between consecutive chunks often yield inconsistent predictions, degrading temporal coherence and the task success rate. Existing methods, such as inference-time blending, merely reweight mismatched proposals without correcting underlying errors, leading to residual accumulation under biased or noisy histories. We propose ChunkFlow, a seam-aware training-and-execution framework for chunked policies that aligns chunk structure with boundary execution. It partitions each chunk into frozen, editable, and future zones, applies deterministic overlap blending at execution, and trains raw predictions with seam and first- and second-order continuity losses. History corruption and scheduled sampling improve robustness to executed-history errors, while an AWAC fine-tuning stage adapts the policy without removing these structural regularizers. Under mild smoothness assumptions, pre-blending seam discrepancies provably decay with increasing overlap. Experiments on CALVIN, LIBERO, and real robots show an improved success-stability trade-off with low-latency inference. Project page: https://cytoderm-ai.github.io/chunkflow.

cs.RO↗
arXiv2026

AppAgent: Multimodal Agents as Smartphone Users

Recent advancements in large language models (LLMs) have led to the creation of intelligent agents capable of performing complex tasks. This paper introduces a novel LLM-based multimodal agent framework designed to operate smartphone applications. Our framework enables the agent to operate smartphone applications through a simplified action space, mimicking human-like interactions such as tapping and swiping. This novel approach bypasses the need for system back-end access, thereby broadening its applicability across diverse apps. Central to our agent's functionality is its innovative learning method. The agent learns to navigate and use new apps either through autonomous exploration or by observing human demonstrations. This process generates a knowledge base that the agent refers to for executing complex tasks across different applications. To demonstrate the practicality of our agent, we conducted extensive testing over 50 tasks in 10 different applications, including social media, email, maps, shopping, and sophisticated image editing tools. The results affirm our agent's proficiency in handling a diverse array of high-level tasks.

cs.CV↗
arXiv2026

Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation

While recent work in Reinforcement Learning with Verifiable Rewards (RLVR) has shown that a small subset of critical tokens disproportionately drives reasoning gains, an analogous token-level understanding of On-Policy Distillation (OPD) remains largely unexplored. In this work, we investigate high-loss tokens, a token type that--as the most direct signal of student-teacher mismatch under OPD's per-token KL objective--should progressively diminish as training converges according to existing studies; however, our empirical analysis shows otherwise. Even after OPD training reaches apparent saturation, a substantial subset of tokens continues to exhibit persistently high loss; these tokens, which we term Rock Tokens, can account for up to 18\% of the tokens in generated outputs. Our investigation reveals two startling paradoxes. First, despite their high occurrence frequency providing a disproportionately large share of total gradient norms, Rock Tokens themselves remain stagnant throughout training, resisting teacher-driven corrections. Second, through causal intervention, we find that these tokens provide negligible functional contribution to the model's actual reasoning performance. These findings suggest that a vast amount of optimization bandwidth is spent on structural and discourse residuals that the student model cannot or need not internalize. By deconstructing these dynamics, we demonstrate that strategically bypassing these ``stumbling blocks'' can significantly streamline the alignment process, challenging the necessity of uniform token weighting and offering a more efficient paradigm for large-scale model distillation.

cs.CL↗
arXiv2026

Modularized Reinforcement Learning on LLMs: From MDP Creation to Exploration and Learning

Reinforcement learning (RL) has become central to LLM post-training, yet the methods that dominate current pipelines, PPO and GRPO, represent only a narrow slice of what RL offers. Understanding why these methods prevail, and what alternatives exist, requires a principled examination of the design decisions that underlie any RL algorithm. This survey organizes that examination around three stages of algorithm construction. We begin with MDP creation: how the reward function, state space, action space, termination condition, and discount factor are, or could be, defined for LLM training. We then turn to exploration, covering temperature sampling, entropy regularization, intrinsic motivation, tree search, and curriculum learning. Finally, we address learning along four classical RL dimensions: model-free versus model-based, value-based versus policy-based versus actor-critic, on-policy versus off-policy, and credit assignment, including both Monte Carlo methods, which rely on full return estimates, and bootstrapping methods, which update estimates using other learned predictions. Mapping the LLM literature onto this taxonomy reveals a strikingly non-uniform distribution of research effort. Critic-free policy gradients and Monte Carlo credit assignment are densely populated, while value-based methods, off-policy actor-critic training, and bootstrapping-based credit assignment remain largely unexplored despite well-established counterparts in classical RL. These gaps represent concrete opportunities for transferring proven RL techniques to LLM training. By making these gaps explicit alongside the methods that have proven effective, this survey offers researchers in both RL and LLMs a shared framework for understanding current practice and identifying promising directions for future work.

cs.LG↗
arXiv2026

VeriGraph: Towards Verifiable Data-Analytic Agents

LLM-based agents have demonstrated strong capabilities in data-intensive analytical tasks, yet their outputs are rarely verifiable: a reliance on linear text trajectories makes their reasoning difficult to audit. In particular, deterministic computations over raw data and semantic deductions over natural-language claims are often entangled in an unstructured stream, leaving numerical conclusions hard to reproduce and qualitative judgments hard to inspect. To address this, we propose VeriGraph, a traceable neuro-symbolic reasoning framework that enables agents to construct an explicit heterogeneous evidence directed acyclic graph (DAG) during execution. VeriGraph introduces three evidence-expansion primitives, namely computational, grounding, and derivational expansion, to connect raw data, interpreter variables, computed results, and natural-language claims in a unified graph. Under this formulation, structural traceability is reduced to graph reachability from raw data sources to terminal claims, while semantic support is measured by claim-level evidence evaluation. To improve graph construction, we further design a graph-based policy optimization strategy with a composite reward that jointly supervises answer correctness, computational integrity, and derivational coherence. Experiments on four benchmarks show that VeriGraph-8B achieves the highest overall score among all baselines. More importantly, VeriGraph produces auditable evidence graphs with substantially stronger claim grounding, achieving a 87.61\% Grounding Rate under our claim-level evidence support evaluation. These results suggest that explicit evidence-graph construction is a promising path toward verifiable data-analytic agents. Our code is available at https://github.com/ignorejjj/VeriGraph.

cs.CL↗
arXiv2026

Biological Reasoning-Informed Regression for Interpretable Regulatory DNA Activity Prediction

DNA cis-regulatory elements (CREs) such as enhancers control gene expression levels. Accurately predicting regulatory activity from DNA sequences is valuable but challenging, as it requires understanding complex biological regulatory processes. Existing methods typically regress activity scores from sequences in a black-box manner, limiting both interpretability and regression performance. Meanwhile, large language models (LLMs) benefit from explicit reasoning processes, yet directly applying LLMs to raw DNA sequences performs poorly. In this paper, we bridge this gap by introducing R3LM, a framework that teaches LLMs reasoning-informed regression on regulatory DNA through structured biological knowledge. Specifically, we design a biologically grounded data format that structures DNA's regulatory information for improved LLM understanding, and construct CRE-ReasonBench, the first dataset that associates DNA sequences and activity scores with mechanistic reasoning traces. Through two-stage training that first teaches LLMs reasoning over structured biological information then performs regression, R3LM achieves state-of-the-art performance on enhancer prediction across three cell types, outperforming both LLMs with raw sequence input and specialized DNA models while providing interpretable mechanistic explanations. We expect R3LM as an interpretable reward model that can effectively assist biologists in CRE design. Code is available at https://github.com/DuanYi516/R3LM.

q-bio.GN↗
arXiv2026

Reinforcement Learning from Cross-domain Videos with Video Prediction Model

Reinforcement learning from expert videos across visually distinct domains is challenging due to the absence of reward signals and the presence of domain gaps. We introduce XIPER (Cross-domain Video Prediction Reward), a reward model for learning from expert videos collected in a visually different domain, where the agent's appearance differs due to factors such as color, morphology, or the sim-to-real gap. More specifically, XIPER trains a cross-domain video prediction model that maps agent observations into the expert domain and uses the prediction likelihood as a reward signal. Experiments on the DMC Color Suite (8 tasks) and DMC Body Suite (3 tasks) show that XIPER consistently outperforms baselines despite domain gaps such as differences in agent color and morphology. We further analyze XIPER on a sim-to-real transfer dataset, demonstrating that it produces meaningful reward signals for real-robot observations given only simulated expert videos. Code, pretrained models, datasets and video demonstrations can be found on our project webpage: https://sites.google.com/view/xiper

cs.CV↗