arXiv subjects

Yang Li

Publications and source records attributed to Yang Li.

At least 37 records · Page 2Linked to original sources

arXiv2026

What Should Robot Policies Learn to Predict? Rethinking Supervision Targets in Vision-Language-Action Learning

Vision-language-action (VLA) policies are typically trained with action prediction as the primary supervision signal, since robots ultimately execute actions. Auxiliary objectives such as latent-action (LA) and future-vision (FV) prediction are often added to improve policy learning, yet they compete with action prediction for the same training budget. We ask whether, under a fixed budget, allocating a substantial share of training samples to such objectives, rather than to further action prediction, can improve policy learning. We build a unified VLA framework that couples a two-stream multimodal diffusion transformer (MMDiT) with a sample-level router, which assigns each training sample to either action prediction or an auxiliary objective and thus makes supervision allocation explicit. Within this framework, we instantiate three LA targets based on LAPA and its DINO-feature variants, and three FV targets based on DINOv3, V-JEPA 2, and Wan2.2 representations. Across LIBERO, CALVIN, and four real-robot tasks, suitable LA and FV supervision outperforms action-only training under the same budget, even when action prediction receives no more than half of the training samples. FV generally outperforms LA, although the strongest LA target remains competitive on CALVIN. Diagnostic analyses suggest that DINO-based LA targets emphasize relative motion, whereas FV targets emphasize absolute position and task identity. We therefore argue that what robot policies learn to predict, and how training is allocated among these targets, should be treated as explicit design choices rather than defaulting to action-dominant training.

cs.RO↗

arXiv2026

Source-Aware Dynamic Pricing with Strategic Buyers

In modern commerce, firms and platforms typically use customer features from different sources to personalize prices: some features are verified using seller records or third-party data, while others are self-reported and may be strategically manipulated by buyers seeking lower prices. We therefore introduce a novel source-aware framework for contextual dynamic pricing with strategic buyers in a high-dimensional semiparametric setting, where buyer valuations follow a sparse linear model with an unknown noise distribution. To account for differences in manipulability across feature sources while enabling effective valuation learning, we develop the Source-Aware Localized exploration and calibrated Exploitation (SALE) pricing policy, which draws on verified features to localize price exploration and combines verified and reported features to correct for strategic manipulation during exploitation. Our localized exploration design allows converting binary purchase feedback into unbiased signals of conditional mean valuations, yielding a new population equation that supports consistent sparse estimation despite non-i.i.d. buyer arrivals. Building on this estimate, we develop a localized empirical-process analysis that exploits intrinsic sparse geometry to establish uniform guarantees for nonparametric estimation of the noise distribution in the high-dimensional setting. With these estimation guarantees, we establish SALE's self-recovery property, which keeps localized exploration valid with high probability and leads to a sublinear expected regret bound. Both simulation studies and real-data experiments demonstrate the appealing performance of SALE. To the best of our knowledge, our study is the first to provide a theoretical and practical framework for incorporating information provenance into dynamic pricing with strategic buyers.

stat.ML↗

arXiv2026

Context Value Informed In-Context Reinforcement Learning

In-context reinforcement learning (ICRL) has emerged as an effective paradigm for test time adaptation to unseen tasks without parameter updates. However, existing ICRL methods can exhibit brittle and unstable adaptions, and the mechanisms underlying such adaptions remain poorly understood. We provide a policy-gradient view of ICRL and argue that relying on trajectory-level return feedback can lead to high-variance updates when test time rollout is limited. Motivated by the actor--critic principle, we propose Context Value Informed ICRL (CV-ICRL), which equips the in-context policy with an explicit context value: the expected discounted return conditioned on the current state and accumulated interaction context. CV-ICRL trains a value head and writes its prediction back into the context as a value token, enabling TD-style return targets for lower-variance policy updates. Experiments on the Dark Room, Minigrid, and Procgen testbeds show that CV-ICRL substantially improves the stability of test time adaptation and achieves higher returns across tasks and environments. Code and data are available at https://github.com/Bluixe/CV-ICRL .

cs.LG↗

arXiv2026

TGPO: Trace-Guided Policy Optimization for Robot Task Planning via Verifiable Subgoal Generation

Robot task planning in real-world environments requires mapping abstract natural language instructions to executable action sequences under long horizons and complex constraints. While large language models (LLMs) provide strong commonsense reasoning, they often fail to generate reliable and feasible plans. In contrast, symbolic planners ensure feasibility and optimality but require well-specified goals and cannot directly interpret high-level human intent.We formulate robot task planning as learning to generate verifiable subgoals in the Planning Domain Definition Language (PDDL), bridging language understanding and symbolic planning. To address the challenges of sparse and noisy supervision, we propose Trace-Guided Policy Optimization (TGPO), a reinforcement learning framework that improves structured subgoal generation through (i) verifier-grounded rewards, (ii) external correction of intermediate reasoning traces, and (iii) constrained policy updates that incorporate corrected traces into training.We evaluate TGPO on large-scale household planning tasks with long horizons, abstract instructions, and complex constraints. TGPO significantly outperforms prompting-based and reinforcement learning baselines, with the largest gains on abstract tasks. Furthermore, TGPO integrates naturally with symbolic planners and language-conditioned executors, enabling robust long-horizon planning and combinatorial generalization in realistic environments.

cs.RO↗

arXiv2026

SecJev: Bringing Security Expertise to System One Decision Models

Security workflows need models that turn complex observations and explicit policies into decisions. System One models introduced by Jev return typed predictions and probabilities; security specialization supplies the domain expertise behind those predictions. We introduce SecJev, to our knowledge the first family of Jev-style decision models specialized for security, spanning 0.8B to 9B parameters. Built on Kev's single-pass candidate scorer, SecJev learns Boolean, choice, and ordered decisions from text, telemetry, and observation histories. We develop SecJev-Corpus to unify source-label prediction and explicit-policy evaluation across 14 tasks and eight sources. It covers tool outputs, traffic, federated updates, consensus, authentication, and vehicle messages. Scene-weighted training adapts the models across these domains while preserving a shared typed decision interface. Security specialization improves every model in the family; SecJev-0.8B outperforms general Kev-9B by 20.51 percentage points in task-macro accuracy. At 0.8B and 4B, SecJev achieves accuracy and latency close to answer-only generative fine-tuning with lower peak allocated GPU memory. SecJev-4B improves on Kev-4B and security-trained generation in prompt-injection and traffic transfer tests. We release adapters, decision heads, SecJev-Corpus, and training and inference code.

cs.CR↗

arXiv2026

LabCompiler: A Framework for Verifiable Human-Machine Laboratory Execution from Imperfect Protocols

Closing the loop of AI-driven science requires reliably translating natural-language protocols into executable laboratory workflows despite protocol defects and heterogeneous hardware. Existing approaches often couple scientific interpretation with device-specific code generation, making errors difficult to localize and limiting reuse across laboratory platforms. We introduce LabCompiler, a robust protocol-to-execution framework that transforms imperfect natural-language protocols into verifiable human-machine workflows. Its protocol firewall diagnoses and repairs defects before compilation while preserving unaffected content, guided by troubleshooting knowledge accumulated through perturbation-guided self-evolution. A two-level, device-agnostic graph representation separates experimental semantics from hardware-specific implementation and supports capability-based device assignment with coordinated human execution. We further introduce a protocol recovery benchmark covering four controlled perturbation types and naturally occurring errors collected from protocols written by iGEM participants. Experiments show that LabCompiler outperforms frontier LLMs in protocol error correction while enabling coordinated human-machine execution across heterogeneous, partially automated laboratory settings.

cs.AI↗

arXiv2026

MUCO: A Mani-Unit-Structured Manipulation Framework for Compositional Long-Horizon Robot Tasks

Learning and executing long-horizon robot manipulation as complete task trajectories raises data-collection costs, lim?its capability reuse, and causes local failures to cascade into later task stages. We introduce manipulation units (MUs) as reusable operations, each defined by a local goal, an initi?ation condition, and observable success and failure criteria. MUs are learned and verified individually. During execution, consecutive MUs remain coupled through shared environ?ment dynamics. We formulate MU-Structured Manipulation (MUSM) and realize it through MU-Centric Orchestration (MUCO). Given language and vision, MUCO constructs an MU sequence, matches each MU to compatible policy models in a frozen capability registry, and dispatches the model with the highest calibrated success rate. Visual verification checks MU outcomes against success and failure criteria and governs transitions, enabling retry or fallback to another model from the current state without re-executing verified MUs. Canonical keys enable checkpoint reuse for equivalent MUs across tasks without new demonstrations or updates. We evaluate MUCO on RLBench, RoboTwin 2.0, and five real-world tasks. MUCO achieves higher complete-task success and verified progress, stronger failure containment, and lower error propagation than baselines. Results demonstrate cross-task MU reuse and het?erogeneous policy composition without joint training, shared backbones, or common action representations.

cs.RO↗

arXiv2026

Proofs and disproofs of some conjectures on zero-sum problems of new types

For integers $r\ge1$ and $n\ge2$, let $s_r(n)$ be the least $N$ such that every sequence of $N$ vectors in $\Z^r$, each nonzero modulo $n$, contains $n$ terms whose sum is zero modulo $n$ but nonzero modulo $n^2$. Let $w_r(n)$ be the corresponding least $N$ when any nonempty subsequence is allowed. These constants were introduced as zero-sum problems of new types by Z.-W. Sun. We prove $s_r(n)\ge(2^{r+1}-2)(n-1)+1$ by constructing an explicit sequence from a lattice polytope, \emph{disproving} his conjecture $s_2(n)=4n+1$ for $n\ge4$. We obtain $w_r(2)=2r+1$ by classifying all sequences avoiding such nonempty subsequences modulo $4$. We also \emph{prove} Sun's conjecture $w_2(n)=4n-3$ for $n=2^k3^\ell p$, where $p$ is a prime and $k,\ell\ge0$ are integers, by classifying the extremal sequences of length $4p-4$ modulo $p^2$ when $n=p$. For every fixed $n\ge2$, we prove $w_r(n)/s(\Z_n^r)=O_n(r^2/2^r)\to0$ as $r\to\infty$, where $s(\Z_n^r)$ is the classical Erdős-Ginzburg-Ziv constant, with an explicit threshold on $r$ for strict inequality. Thus, for every fixed modulus $n\ge2$, Sun's proposed equality fails for sufficiently large $r$.

math.CO↗

arXiv2026

Compact deterministic liquid-crystal polarization controller with single-measurement direct control

Polarization controllers (PCs) and their control algorithms are critical for compensating environmental disturbances and maintaining signal fidelity in polarization-sensitive photonic systems. However, existing controllers often involve trade-offs among key performance metrics,while associated control algorithms either converge inefficiently or require multiple state-of-polarization (SOP) measurements. This work presents a compact deterministic liquid-crystal polarization controller (LC-PC) with balanced overall performance, featuring low insertion loss (<1 dB), low driving voltage (<6 V), and full Poincaré sphere coverage. The device exhibits highly predictable and repeatable polarization modulation characterized by SOP-independent trajectory normal vectors, enabling a single-measurement direct-calculation algorithm using pre-characterized rotation axes and voltage-to-rotation relationships, with degree-level polarization control accuracy. The combination of this compact deterministic LC-PC and its efficient control scheme provides a practical solution for polarization management in quantum key distribution, optical sensing, and advanced photonic systems.

quant-ph↗

arXiv2026

KiToke: Distribution-Aware Token Compression for Video Large Language Models

Video Large Language Models (Video LLMs) achieve strong performance on video understanding tasks but suffer from high inference costs due to the large number of visual tokens. We study training-free, query-agnostic video token compression, where heuristic importance rules can over-represent large redundant visual groups and miss less frequent visual patterns, particularly under aggressive token budgets. We propose KiToke, a distribution-aware compression framework that treats visual tokens as samples from the video's spatiotemporal token distribution. KiToke uses global embedding density as a proxy for representational redundancy and inverse-density sampling to balance token allocation across visual patterns of different frequencies. It further aggregates unselected tokens into retained anchors within detected temporal intervals to reduce cross-transition mixing. Across evaluated backbones, KiToke achieves the highest average accuracy among compared training-free methods and remains robust down to 1% retention, where most evaluated baselines underperform uniform random token sampling. On LLaVA-OneVision, it preserves 98.5% and 89.2% of the uncompressed model's four-benchmark mean accuracy at 10% and 1% token retention, with 6.8x and 12.5x prefill speedups, respectively, including compression overhead.

cs.CV↗

arXiv2026

Specification-Guided Synthesis of Deadlock-Free Communication Protocol Refinements with Large Language Models

Ensuring behavioural correctness in communication protocols is a central challenge in distributed software systems, as subtle inconsistencies can lead to deadlocks. In such settings, protocol refinement - the safe substitution of a protocol that preserves correctness and compatibility with other components - is essential. Large language models (LLMs) have demonstrated strong capabilities in code generation and program synthesis, yet lack mechanisms to reliably produce outputs with correct behaviour. Formal specification approaches, such as multiparty session types (MPST), offer rigorous guarantees, including deadlock freedom, but provide limited support for automatically constructing protocol refinements. In this paper, we present Syntropy, a framework for synthesising protocol refinements guided by MPST specifications and LLMs. It incorporates refinement constraints directly into the generation process, ensuring the generated variants satisfy these guarantees. Our comprehensive evaluation indicates that Syntropy achieves 95.6%-99.5% validity while maintaining high syntactic correctness, and produces diverse, non-trivial refinements across multiple LLMs.

cs.SE↗

arXiv2026

WorldRecon: Generative Multi-View Mesh Reconstruction for Large-Scale 3D World

Reconstructing complete, high-quality meshes of large-scale scenes from unposed multi-view images requires faithful geometric grounding while plausibly completing surfaces that are unobserved or only sparsely captured. We present WorldRecon, a generative multi-view reconstruction framework for recovering complete, large-scale 3D scene meshes from unposed RGB images. WorldRecon uses a coarse, noisy global point cloud and camera poses estimated from the input images by a reconstruction model as a global scaffold for whole-scene generation and partitions it into overlapping spatial blocks. To robustly leverage these imperfect geometric priors, we equip the native 3D generative model TRELLIS.2 with a dedicated reconstruction-based conditioning architecture that generates each block from its local point-cloud crop and block-relevant image observations, producing sparse structure, detailed geometry, and PBR materials. To maintain scene-level coherence, we introduce neighbour-aware latent inpainting and a causal world-generation procedure that propagates structural, geometric, and material context across overlapping blocks. Together, these designs scale multi-view generative reconstruction to large-scale scenes, yielding coherent, unified meshes. Extensive experiments demonstrate that WorldRecon achieves state-of-the-art reconstruction performance, reducing Chamfer distance by 14.9% and 43.1% while improving F-score by 17.3% and 12.0% over the strongest baselines on synthetic and real-world indoor scenes, respectively, thereby demonstrating faithful and complete generative reconstruction as well as strong sim-to-real generalization. Project page: https://kszpxxzmc.github.io/WorldRecon-website/.

cs.CV↗

arXiv2026

TokenPB: Enabling VLA Action Self-Verification via VLM Token Probability Bucketing

Test-time sampling can improve Vision-Language-Action (VLA) policies, but only if the model can efficiently select a good action from multiple candidates. Prior work often relies on separately trained external verifiers, adding extra models and inference overhead. We propose \textbf{Token Probability Bucketing (TokenPB)}, a self-verification approach that discretizes continuous action chunks into bucket tokens and co-trains a single VLA checkpoint with a joint objective: a flow-matching loss for action generation and an auxiliary bucket loss $\mathcal{L}_{\text{bucket}}$ that teaches the VLM backbone to model observation-conditioned distributions over bucket-token sequences. At inference, TokenPB samples $M$ candidates, scores each via teacher-forced evaluation under the learned distribution, and selects the best---enabling efficient batched scoring with observation-prefix KV-cache reuse. We provide theory connecting the training objective to a monotonic ranking property and bounds on expected selection regret that separate candidate-set effects from calibration. Across multiple backbones and benchmarks, TokenPB yields consistent success-rate improvements over single-sample inference from the same checkpoint (+2.7 in simulation, +17.1 in real-world, +16.4 under out-of-distribution shifts), and further improves over external-verifier baselines under the same sampling budget. The code is publicly available at https://github.com/RUCKBReasoning/TokenPB.

cs.CV↗

arXiv2026

On the Sparsity-Storage-Accuracy Tradeoff in Parsimoniously Activated Dictionary Learning

Dictionary learning has long been studied from both optimization and probabilistic perspectives. While formulations with element-wise sparsity regularization (e.g., L1-based sparse coding) admit well-established probabilistic interpretations, many structured variants that impose global constraints lack a clear and tractable generative view. In this paper, we revisit a class of practically effective yet theoretically under-explored dictionary learning methods that impose a simple global regularization on the number of activated dictionary atoms, which we term parsimoniously activated dictionary learning (PADL). We show that PADL admits an equivalent formulation as maximum a posteriori estimation under a structured generative model, with auxiliary latent variables that govern global activation patterns. This formulation allows us to derive generalization guarantees that are difficult to obtain under the original formulation. More importantly, it yields an analytical characterization of the tradeoff between sparsity, storage cost, and reconstruction accuracy, enabling data-driven estimation of optimal hyperparameters. Based on this connection, we develop an efficient and interpretable PADL algorithm that eliminates manual hyperparameter tuning, achieving improved reconstruction performance under comparable sparsity levels on visual benchmarks. We further demonstrate its practical utility in accelerating inference for vision-language models.

cs.LG↗

arXiv2026

StepAudio 3 Realtime Technical Report

Realtime spoken interaction demands deep reasoning, prompt responses, and fluid turn-taking. We present StepAudio 3 Realtime, an audio-language foundation model organized around a continuous listen-converse-think-act loop. Deep Perception captures rich acoustic cues to interpret user intent, while Seamless Duplex models synchronized audio streams to handle pauses, backchannels, and interruptions naturally. Crucially, we resolve the tension between deep deliberation and latency via Think-While-Speaking, executing private reasoning in parallel with spoken delivery. In reasoning mode, StepAudio 3 reaches a 73.0 macro average on StepAudioChat. With Think-While-Speaking, it achieves dialogue and reasoning performance comparable to dedicated reasoning models while speaking in real time. Furthermore, an integrated Voice Agent handles asynchronous tool execution without disrupting the dialogue flow. StepAudio 3 Realtime achieves top-tier performance across key dimensions: an exceptional 90.6 on the MMSU benchmark, 98.9 Overall on the Artificial Analysis Full-Duplex Bench, and a 56.0% macro task-success rate on $τ$-Voice.

cs.SD↗

arXiv2026

Embodied Skill Bench: Benchmarking How Well Agents with Embodied Skills Perform on Physical Tasks

LLM agents increasingly control robots using VLA policies, perception models, and procedural knowledge. These modules are difficult to reuse because their interfaces, embodiment assumptions, and conditions of use differ, and their benefits on physical tasks remain unclear. We package them as embodied skills that extend agent skills with explicit embodiment assumptions and conditions of use, building 250 action, perception, and experience skills from 1,052 resources. On this basis we introduce EmbodiedSkillBench, which runs an agent in a closed observation-action loop and records success, actions, tokens, and wall-clock time. It supports three modes: native control, skill-only execution, and Agent-Directed Reuse (ADR), a mode we propose in which the agent keeps its native controls and decides for each skill whether to invoke, modify, or ignore it. Across four agents on 100 perturbed LIBERO-Pro tasks, we find that skills help an agent only as far as they exceed what it can already do. Skill-only execution lifts GPT-5.6-Sol from 11% to 63% and Kimi-K3 and Claude-Opus-5 by about 20 points, while cutting their token usage by 58% to 85%. Yet all four agents then end up between 61% and 64%, and GPT-6-Astra falls from 92% under native control to 61%. This suggests that as agents grow stronger, the role of embodied skills shifts from supplying missing capability to reducing cost: GPT-6-Astra already solves most tasks through native control, so restricting it to skills removes more than it adds. Under ADR, which lets it keep native control and call skills selectively, GPT-6-Astra recovers to 85%, within 7 points of native control, while using 14.1% fewer tokens and 11.9% less time. With Jev as a skill choosing motion directions and gripper actions, it reaches 81% with 50% fewer tokens and 57% less time. The code is available at https://github.com/HanZhi306/EmbodiedSkill-Bench.

cs.RO↗

arXiv2026

EmbodiedHead: Real-Time Listening and Speaking Avatar for Conversational Agents

We present EmbodiedHead, a speech-driven talking head framework that equips LLMs with real-time visual avatars for conversation. A practical embodied avatar must achieve real-time generation, unified listening-speaking behavior, and high rendered visual quality simultaneously. Our framework couples the first Rectified-Flow Diffusion Transformer (DiT) for this task with a differentiable renderer, enabling diverse, high-fidelity generation in as few as four sampling steps. Prior listening-speaking methods rely on dual-stream audio, introducing an interlocutor look-ahead dependency incompatible with causal user--LLM interaction. We instead adopt a single-stream interface with explicit per-frame listening-speaking state conditioning and a Streaming Audio Scheduler, suppressing spurious mouth motion during listening while enabling seamless turn-taking. A two-stage training scheme of coefficient-space pretraining and joint image-domain refinement further closes the gap between motion-level supervision and rendered quality. Extensive experiments demonstrate state-of-the-art visual quality and motion fidelity in both speaking and listening scenarios. Project page: https://vpx-ecnu.github.io/EmbodiedHead-Website/.

cs.CV↗

arXiv2026

Agent-UCT: Upper Confidence Bounds Applied to Trees for Cost-Aware Agentic Workflow Optimization

Workflow search pays for complete quality observations, whose execution cost changes as reusable prefixes accumulate. AgentUCT uses an economic model of the remaining evaluation cost to guide UCT selection while backing up raw quality. Query-prefix accounting connects the search decisions to reusable execution state. For finite fixed-reward trees with bounded adaptive cold-cost estimates and successful persistent caching, we prove an $O(\log T)$ bound on cumulative quality regret over $T$ complete evaluations. Across five paired seeds, AgentUCT achieves 69.57% test F1 versus 68.01% for UCT at a 32M weighted-token budget on a retrospective HotpotQA panel. On GSM8K, it reaches 93.07% mean accuracy at the first budget, compared with 92.60% for UCT and 91.60% for AFlow. Single-factor RAG controls show that online prices and historical continuations improve the selected workflows at equal retained evaluation counts.

cs.AI↗