Searcharxiv⌕ Search

arXiv subjects

Yixuan Li

Publications and source records attributed to Yixuan Li.

At least 19 recordsLinked to original sources
arXiv2026

SpecCoder: Specification-Aware Code Generation with Curriculum Dual-Task Reinforcement Learning

Large language models (LLMs) have made substantial progress in code generation but still struggle with challenging programming tasks that require understanding rich natural language requirements. These requirements often specify problem goals, input/output formats, constraints, examples, and edge cases. Overlooking even one may produce executable but functionally incorrect code. Existing training-free methods mainly rely on prompting or agent-based workflows, while training-based methods typically optimize final code outputs. However, existing approaches provide limited supervision for learning the intermediate mapping from raw requirements to structured specifications and for grounding them in concrete implementation behavior. Consequently, models may omit critical constraints, and even when an explicit specification is produced, the implementation may fail to reflect it consistently. Motivated by this gap, we propose SpecCoder, a specification-aware two-stage training framework for code generation. SpecCoder first employs specification-guided SFT to train LLMs to derive structured specification analyses and generate code conditioned on them. It then introduces curriculum dual-task GRPO, which jointly optimizes specification-guided generation and discrimination to encourage stronger correspondence between specifications and code behavior. Experiments on APPS, CodeContests, and xCodeEval demonstrate the effectiveness of specification-aware training, with SpecCoder consistently improving both standalone code generation and agent-based workflows. Additional evaluations on BigCodeBench-Hard and ClassEval, alongside human evaluation and perturbation studies, further validate the role of structured specifications in guiding code generation and discrimination.

cs.SE↗
arXiv2026

Embodied Scene Rearrangement Planning

This paper introduces Embodied Scene Rearrangement Planning (ESRP), a novel task requiring embodied agents to rearrange furniture in 3D scenes to match a target configuration using only egocentric observations and a top-down target layout. Unlike prior rearrangement tasks, ESRP precludes global state access and introduces mutual object occlusions, reflecting the practical constraints of real-world robotic deployment. These factors make aligning partial egocentric observations with the global target layout particularly challenging for long-horizon planning. To facilitate research, we present ESRP-Bench, a comprehensive benchmark built on OmniGibson featuring over 5,400 scene pairs and 8,200 objects. We define three multi-level metrics to evaluate rearrangement quality and provide four baselines: a hierarchical task-and-motion planning method, a vision-language-model-based method, and two learning-based approaches (IL and RL). Experimental results demonstrate that current methods struggle to complete the task efficiently, highlighting ESRP as a challenging frontier for embodied agents in scene understanding and long-horizon task planning. This work serves as a stepping stone toward deploying intelligent agents in real-world scenarios. Project page: https://pie-lab.cn/ESRP/.

cs.RO↗
arXiv2026

FireRedAudio: A General-Purpose Audio Language Model with Decoupled Continuous Representations for Understanding and Generation

A unified audio model must recognize and understand linguistic, paralinguistic, and environmental information while supporting speech synthesis and editing. A key challenge is representation: understanding favors compact features suited to long-context modeling, whereas speech generation requires reconstructible features that preserve fine-grained acoustic detail. We introduce FireRedAudio, a general-purpose audio language model with a shared 9B-parameter LLM. To the best of our knowledge, it is the first publicly disclosed unified audio-language model to provide separate continuous input representations for understanding and generation within a single trainable autoregressive LLM. Audio to be recognized or analyzed is processed by a dedicated Audio Encoder, while speech inputs for generation use a RedAE-based pathway. The LLM directly generates text or conditions a flow-matching DiT to produce continuous acoustic latents. Through progressive multitask training, FireRedAudio supports ASR and audio understanding, with the latter extending to recordings of up to one hour, as well as zero-shot TTS, Instruct TTS, and semantic and acoustic speech editing. Its structured organization of long-form audio achieves second-level timestamp accuracy. Across comprehensive evaluations, FireRedAudio achieves competitive or leading performance in audio understanding and multilingual ASR, strong content accuracy and speaker preservation in zero-shot TTS, leading instruction following in Instruct TTS, and substantial improvements over Ming-UniAudio-Edit in both semantic and acoustic speech editing. These results demonstrate the viability of decoupled continuous input representations for unifying audio understanding and continuous-latent speech generation in a model of moderate scale. Our code is available at https://github.com/FireRedTeam/FireRedAudio.

cs.CL↗
arXiv2026

MMAG: A Multi-Control Mixed Audio Generation Benchmark

Recent audio generation systems have progressed from single-modality synthesis to generating complex acoustic scenes containing speech, music, and sound effects. Therefore, evaluating these models requires assessing multiple interacting capabilities, including semantic fidelity, speaker consistency, and temporal control, yet existing benchmarks focus on isolated domains or coarse-grained descriptions. To address this gap, we introduce the Multi-control Mixed Audio Generation (MMAG) benchmark. MMAG contains approximately 4,000 manually verified audio clips with rich annotations covering speech content, speaker identity, music attributes, sound events, and temporal relationships, together with dedicated subsets for voice cloning and timestamp-conditioned generation. We further propose a systematic evaluation protocol that measures acoustic fidelity, speech quality, semantic alignment, and temporal accuracy. Benchmarking representative agentic orchestrators, unified audio-visual generation models, and native mixed-audio generators reveals substantial performance trade-offs across these capabilities, with no existing model performing consistently well. Our results highlight the remaining challenges of controllable mixed audio generation and establish MMAG as a comprehensive benchmark for future research.

cs.SD↗
arXiv2026

Patients-like-me: A Variational LM--GNN Framework for Explainable Clinical Prediction

Language models (LMs) offer strong textual representations for electronic health records (EHRs), but they encode patient sequences in isolation and provide limited explainability. Graph neural networks (GNNs) complement LMs by incorporating inter-patient relationships and enabling reference-patient attribution, yet they rely on high-quality patient representations. We propose Patients-like-me (PLM), a unified LM--GNN framework that integrates local patient semantics with global cohort structure. To train PLM efficiently, we introduce a Variational Expectation-Maximization algorithm that alternates LM and GNN updates under a supervised variational objective. Extensive experiments on MIMIC-III and MIMIC-IV show that PLM consistently outperforms state-of-the-art methods, with improvements generalizing across encoder-only and decoder-only LM backbones. These gains are achieved with only modest additional computational overhead. PLM also provides reference-patient explanations by retrieving influential similar patients, while edge-masking experiments confirm that the highest-ranked references have the greatest impact on model predictions.

cs.CL↗
arXiv2026

CameraAnything: Refilming Videos with Arbitrary Camera Control

We introduce CameraAnything, the first unified framework for camera controlled video editing that enables joint control of both intrinsic and extrinsic camera parameters. Existing approaches either rely on expensive 3D reconstruction to achieve full camera functionality or restrict editing to extrinsic parameter manipulation. Moreover, the coupled influence of intrinsic and extrinsic parameters on video appearance makes disentangled modeling particularly challenging. To address this, we adopt per-pixel Pl\"ucker ray injection alongside resolution-aware 3D RoPE in self-attention, building both camera conditioning and spatial positional encoding on the target latent to jointly control camera position, focal length, and native resolution editing without cropping or outpainting. To overcome the scarcity of paired training data, we further develop a scalable synthetic pipeline that constructs diverse dynamic scenes through structured multi-camera recording and generates synchronized videos with varied camera configurations. With a tailored orthogonal training strategy, CameraAnything enables expressive video reshooting with arbitrary viewpoint control, focal length adjustment, resolution adaptation, and multi-shot transitions within a single generation process, offering strong practical value for cinematic video editing and cross-platform content adaptation in video production.

cs.CV↗
arXiv2026

SLVMBench: Skill Learning from Video Memory

We introduce Skill Learning from Video Memory (SLVMBench), the first benchmark that jointly evaluates whether video large language models (video-LLMs) can learn skills from long video memory and apply them to real-time tasks. SLVMBench presents models with 2-3 hour video streams that contain a tutorial video embedded in a stream of arbitrary irrelevant videos, resembling real-world human learning practices. Video-LLMs are asked to apply the acquired skill to answer real-time questions about an ongoing video. Unlike long-video understanding benchmarks that emphasize passive comprehension and skill-learning benchmarks that rely on short, immediate demonstrations, SLVMBench tests the full pipeline of memorizing and extracting procedural knowledge, as well as transferring it to real-time tasks. Moreover, rigorous human annotations feature sub-second-level temporal calibration, manually engineered questions eliminating common-sense guessing, and collated tutorials to ensure coverage of the required skills. Evaluations on state-of-the-art proprietary and open-source video LLMs show that video-LLMs struggle substantially with learning and applying skill knowledge from videos. Moreover, performance degrades markedly when the skill knowledge is placed within a long video memory. These results reveal a key limitation of existing video LLMs and position SLVMBench as the first benchmark for studying real-time skill acquisition and application from long-context video memory.

cs.CV↗
arXiv2026

WorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory

We present WorldDirector, a highly controllable video world model framework designed for persistent dynamic object memory and unrestricted viewpoint exploration. Unlike existing world models that entangle physical dynamics with pixel rendering and rely on continuous visual observation to sustain motion, our framework explicitly decouples semantic motion orchestration from visual generation. By leveraging an LLM to coordinate 3D trajectories with camera movements and subsequently employing these orchestrated trajectories as control signals for video generation, our approach ensures strict physical logic and appearance stability, successfully preserving the exact visual identities of dynamic entities even when they re-enter the scene after prolonged periods out of view. Experimental results demonstrate that our method supports the synthesis of complex and extended events with unprecedented controllability and persistent dynamic object memory. Project Page: https://worlddirector.github.io/

cs.CV↗
arXiv2026

The Speedup Paradox: Rethinking Inference Speed-Quality Trade-off in Embodied Tasks

Embodied foundation models have recently been widely used to improve robot generalization and task success rates. Previous works apply lossy efficient-inference techniques such as quantization, pruning, and asynchronous inference, accepting small action quality degradation in exchange for lower per-step computation cost and inter-action latency. However, unlike traditional static ML tasks, embodied tasks involve repeated interaction with the environment, and task-level performance is determined not only by per-step cost, but also by closed-loop effects unique to embodied execution, which remain insufficiently characterized in current efficient-inference studies. In this work, we propose TISED (\underline{T}ask-level \underline{I}nference \underline{S}peedup \underline{E}ffect \underline{D}ecomposition), an analytical framework that unifies diverse lossy inference optimization techniques and decomposes their effects on static and dynamic tasks, and uncovers some paradoxical effects on task-level performance: (1) on \textit{static tasks}, optimization sometimes can lengthen end-to-end per-task completion time even as per-step latency drops; (2) on \textit{dynamic tasks}, moderate lossy optimization can raise task success rate even above the baseline; and (3) the monotonicity and sweet-spot location of both effects can shift with hardware configuration. Together, our findings provide a new perspective on adapting inference optimization techniques to embodied tasks.

cs.RO↗
arXiv2026

video-SALMONN-R$^3$: Learning to ReWatch, ReAsk, and ReAnswer for Efficient Video Understanding

Video large language models (LLMs) are often constrained by computation and memory budgets, leading them to use reduced frame rates and spatial resolutions, which may cause them to miss critical information for question answering (QA). A practical and efficient solution is a two-stage paradigm: first perform coarse video understanding to localize relevant segments, and then re-watch these segments at higher temporal or spatial fidelity. In this paper, we present video-SALMONN-R$^3$, the first end-to-end video-LLM that enables re-watch through reinforcement learning without relying on chain-of-thought (CoT) cold-start. This design removes the need for costly CoT data annotations and avoids CoT-based supervised fine-tuning (SFT), which can otherwise degrade the pretrained video understanding abilities. To address the mismatch between the reasoning-first behavior induced by re-watch and the answer-first tendency of pretrained video-LLMs, we propose a re-answer strategy, in which the model first produces a direct answer in the first watch and then refines it after re-watching. Finally, to improve question adherence during re-watching, we propose a re-ask mechanism that re-injects the query when revisiting localized segments. Experimental results show that video-SALMONN-R$^3$ consistently outperforms both the base model and the QA-SFT baseline, while surpassing prior re-watch-based approaches with significantly lower computational cost. Code, models, and data will be publicly released upon acceptance.

cs.CV↗
arXiv2026

ORAgentBench: Can LLM Agents Solve Challenging Operations Research Tasks End to End?

Large language models are increasingly deployed as autonomous agents for multi-step tasks in executable environments, yet their ability to perform realistic operations research (OR) work remains unclear. Existing OR evaluations often decouple modeling from solving, rely on pre-formalized or text-only instances, and rarely test the full workflow from operational artifacts to validated decisions. In this work, we introduce ORAgentBench, an execution-grounded benchmark for evaluating autonomous agents on challenging end-to-end operations research tasks. It contains 107 human-reviewed tasks across diverse operational scenarios, each packaged in an isolated environment with a natural-language brief, multi-file data, configuration artifacts, and a required submission schema. Agents must write and run solution code, and their submissions are evaluated by hidden validators for schema validity, hard-constraint feasibility, and normalized objective quality. Experiments with fourteen frontier agent-model configurations show that current agents remain far from reliable OR practice. The best agent passes only 35.51% of all tasks and 20.59% of hard tasks, and many feasible submissions still fall below the required quality threshold. Failure analysis further shows that errors are dominated by strategic weaknesses, including missed operational rules, brittle formulations, weak feasible-solution construction, and insufficient solution improvement. OR-specific procedural skills increase hard-task feasibility, but do not reliably improve solution quality or pass rate. These results suggest that progress in OR agents requires moving beyond plausible optimization code toward dependable, high-quality operational decision-making.

cs.AI↗
arXiv2026

Coherent Microwave Control of Optically Addressable Donor Qubits in ZnO

Optically addressable shallow donors in ZnO combine efficient spin-selective optical transitions with the potential for long spin coherence in an isotopically purifiable host lattice, making them an attractive platform for spin-photon quantum technologies. A key missing capability, however, has been coherent control beyond the small-angle rotations accessible with ultrafast optical pulses. Here we demonstrate coherent microwave control of implanted $^{115}\mathrm{In}$ donors in ZnO. Resonant optical pumping initializes and reads out the donor electron spin. Pulsed optically-detected magnetic resonance resolves the ten hyperfine transitions associated with the coupled $^{115}\mathrm{In}$ nuclear spin (I = 9/2) and reveals optical-pumping-induced nuclear spin polarization. We observe coherent Rabi oscillations with a maximum Rabi frequency of $\Omega/2\pi = 36.2 \pm 0.7$\;MHz, corresponding to a $\pi$-pulse time of 13.8$\pm$0.3\;ns, and characterize the spin coherence using Ramsey, Hahn echo and dynamical-decoupling measurements. Unexpectedly, the measured coherence is substantially shorter than reported in previous optical studies of donor spins in ZnO at high magnetic field. Control experiments rule out several simple explanations including microwave heating and instantaneous diffusion from the driven donor ensemble, leaving an open question regarding the origin of decoherence at low magnetic field in microwave-controlled ZnO donors. These results establish microwave control of ZnO donor qubits with resonant optical access to specific donor species. More broadly, they demonstrate that coherent microwave control can be achieved in optically addressable spin systems with nanosecond-scale inhomogeneous dephasing, enabling field-, temperature-, and materials-dependent studies of coherence-limiting mechanisms and the development of optically interfaced electron-nuclear spin registers.

quant-ph↗
arXiv2026

Energy Efficiency Optimization for Rotatable Antenna-Enabled Uplink NOMA Systems

This paper investigates a rotatable antenna (RA)-enabled uplink non-orthogonal multiple access (NOMA) system, where a base station equipped with multiple independently RAs serves both ground and aerial users. Specifically, we formulate an energy efficiency (EE) maximization problem by jointly optimizing receive beamforming, user power allocation, and RA rotation. To make the problem tractable, a new block coordinate descent-based algorithm is developed, in which the receive beamforming is updated via the minimum mean square error criterion, while the power allocation and RA rotation are handled by fractional programming and successive convex approximation. Numerical results demonstrate the EE superiority of the proposed RA-NOMA scheme over several benchmarks.

cs.IT↗
arXiv2026

KGEdit: Ambiguity-Aware Knowledge Graphs for Training-Free Precise Video Generation and Editing

In recent years, training-free video generation has progressed remarkably. However, when handling complex textual instructions, existing methods still suffer from semantic ambiguity, incorrect concept binding, and cross-frame inconsistency. To address these issues, we propose KGEdit, a structured semantic control framework for text-to-video (T2V) diffusion models. Specifically, we first construct an ambiguity-aware knowledge graph (AAKG) to disentangle and disambiguate the input prompt, converting it into four types of structured semantics: identity, relation, attribute, and negative constraints. We then design a structured semantic injection module (SSIM) to inject these semantic signals into key layers of the diffusion Transformer, enabling fine-grained semantic control. In addition, we introduce a temporal-aware semantic control (TASC) module that dynamically schedules semantic objectives according to the stage-wise characteristics of the denoising process, further improving semantic alignment and temporal consistency. Experiments show that KGEdit outperforms existing methods in editing precision and temporal stability, while offering higher efficiency and controllability in text-driven interaction scenarios.

cs.CV↗
arXiv2026

UniLab: A Heterogeneous Architecture for Robot RL Beyond GPU-Dominant Paradigms

Simulation-based RL for contemporary robot control is increasingly organized around GPU-resident simulation: physics, rollout collection, and learning are placed on a single GPU-centric execution path. This paradigm has greatly improved training speed, but it has also encouraged a default assumption that efficient training requires physics to reside on the GPU. We revisit this assumption. Our view is that, in simulation-dominated robot control, the essential question is not which processor runs physics, but whether simulation throughput, policy learning, and runtime synchronization form an efficient end-to-end loop. We present UniLab, a heterogeneous CPU-simulation / GPU-learning architecture that decouples CPU-parallel simulation from GPU policy updates through a unified runtime for data movement, buffering, and synchronization. UniLab is implemented as a complete and extensible training system using MuJoCoUni and MotrixSim CPU-batched physics backends, supporting PPO, FastSAC, FlashSAC, and APPO. On representative simulation-based robot control tasks, UniLab improves end-to-end training efficiency by 3--10$\times$ under the same hardware configuration, while reducing dependence on the NVIDIA CUDA-based software stack and supporting cross-platform execution on the Apple macOS platform and the AMD ROCm and Intel XPU accelerator backends. These results show that GPU simulation is an effective path to efficient training, but not a necessary one, broadening the practical system choices available for robot RL training. Project page: https://unilabsim.github.io.

cs.RO↗
arXiv2026

Fast Computational Methods for Regularized Estimating Equations

Estimating equations arise in a wide range of statistical applications, including longitudinal and clustered data analysis, survival analysis, econometrics, and semiparametric inference. In high-dimensional settings, adding sparsity-inducing regularization often leads to computational challenges that are not fully addressed by standard penalized optimization routines. These challenges are closely tied to the structural form of the underlying estimating problem: mainly, the estimating function needs not be the gradient of a scalar objective and may involve asymmetric Jacobians, overidentification, nonsmoothness, nonconvexity, or nested optimization. This article first reviews the application areas of estimating equations, and then the computational methods for regularized estimating equations by organizing them into four broad formulations: minimization-type, Dantzig-type, regularization-type, and fixed-point-type approaches. We discuss the main numerical strategies associated with each formulation, including penalized optimization, constrained linear programming, iterative root-solving, and proximal fixed-point iteration. We also highlight the connection between regularized estimating equations and fixed-point problems, which provides a unified computational perspective for analyzing and solving regularized estimating equations.

stat.CO↗
arXiv2026

OmniMem: Perturbation-aware Memory Compression for Streaming Audio-Visual LLMs

Audio-visual large language models (LLMs) hold strong promise for long-form video understanding, yet their long-video inference is fundamentally limited by the linear growth of video tokens and key-value (KV) caches. We present OmniMem, a memory-efficient streaming framework designed specifically for audio-visual LLMs. Unlike existing compression methods that treat all tokens uniformly, OmniMem introduces a modality-aware memory allocation strategy that separately manages visual and audio contexts, addressing the severe token imbalance between the two modalities. OmniMem further preserves informative and non-redundant KV states through perturbation-aware memory selection, enabling compact memory without sacrificing long-range understanding. To strengthen compression under realistic deployment constraints, we also explore budget-aware fine-tuning, which encourages the model to consolidate useful information into retained memory. Experiments on VideoMME Long, LVBench, and LVOmniBench with video-SALMONN 2+ and Qwen-2.5-Omni show that OmniMem consistently improves over strong training-free compression baselines by 2-4% absolute accuracy under the same memory budgets, with an additional 1-2% gain after fine-tuning.

cs.AI↗
arXiv2026

3DMambaComplete: Exploring Structured State Space Model for Point Cloud Completion

Point cloud completion aims to generate a complete and high-fidelity point cloud from an initially incomplete and low-quality input. A prevalent strategy involves leveraging Transformer-based models to encode global features and facilitate the reconstruction process. However, the adoption of pooling operations to obtain global feature representations often results in the loss of local details within the point cloud. Moreover, the attention mechanism inherent in Transformers introduces additional computational complexity, rendering it challenging to handle long sequences effectively. To address these issues, we propose 3DMambaComplete, a point cloud completion network built on the novel Mamba framework. It comprises three modules: HyperPoint Generation encodes point cloud features using Mamba's selection mechanism and predicts a set of Hyperpoints. A specific offset is estimated, and the down-sampled points become HyperPoints. The HyperPoint Spread module disperses these HyperPoints across different spatial locations to avoid concentration. Finally, a deformation method transforms the 2D mesh representation of HyperPoints into a fine-grained 3D structure for point cloud reconstruction. Extensive experiments conducted on various established benchmarks demonstrate that 3DMambaComplete surpasses state-of-the-art point cloud completion methods, as confirmed by qualitative and quantitative analyses.

cs.CV↗