Skip to main content
Search

Peking University

79 related articles

Subscribe to updates
  1. arXiv

    Refractory high-entropy alloys move past equimolar: a dislocation-genome plus dynamics framework predicts high-temperature strength and room-temperature ductility across the NbMoTaW space

    High-temperature strength and room-temperature ductility in refractory high-entropy alloys have long traded off against each other, and most studies stay near equimolar compositions. This multiscale framework builds energy landscapes for edge-dislocation glide from atomic configurations with a dislocation-genome method, feeds the resulting local configurational forces into stochastic discrete dislocation dynamics, and obtains a linear scaling law between critical shear stress and force fluctuation. That law predicts high-temperature strength across the NbMoTaW composition space, which is then screened jointly with a ductility criterion based on the relative mobility of screw and edge dislocations.

  2. arXiv

    Video anomaly understanding stops relying on pre-sampled frames: SEEK-VAU lets an agent search and verify, lifting evidence F1 from 28.1 to 43.7

    Video anomaly understanding has long used fixed observation: a model receives a bundle of sampled frames or clustered events and then judges the anomaly, with no way to go back for missing evidence once reasoning starts. SEEK-VAU recasts this as budget-bounded agentic search and verification, letting the policy gather evidence across the precursor, trigger and aftermath stages of an event chain and check sufficiency before closing the case. On its SEEK-Bench, built on a Qwen3-VL-8B backbone, it reaches 55.6 Temporal mIoU, 45.5 QA Accuracy and 43.7 Evidence F1, against 44.2, 37.7 and 28.1 for the fixed-observation Qwen3-VL baseline.

  3. arXiv

    Humanoid soccer stops juggling separate controllers: SoccerBot unifies four skills on a human motion prior and transfers zero-shot to a real robot

    Humanoid soccer robots have either covered only a few skills or skipped motion priors entirely and moved unnaturally. SoccerBot distills large-scale human soccer motion-capture data into a structured latent motion prior, letting a single skill-conditioned policy learn moving, dribbling, trapping and kicking in one latent space. On the Unitree G1, the policy transfers zero-shot to hardware, with simulation success rates of 97.04% for kicking and 89.88% for dribbling, while real-robot kicking direction error rises from 4.09 to 10.41 degrees.

  4. arXiv

    Visual verifiers stop answering only true or false: OmniVerifier-C1 splits prompts into sub-questions and lifts ViVerBench overall to 0.684

    Visual verifiers have long returned only a true-or-false verdict, which cannot separate partially correct outputs from worse ones. OmniVerifier-C1 decomposes a complex prompt or environmental rule into fine-grained sub-questions, judges each one binarily, and turns the satisfied fraction into a continuous score, while a decoupled reinforcement-learning scheme optimizes sub-question judgments and global alignment scoring together. On ViVerBench, the Qwen3-VL-8B version trained with graded supervised fine-tuning plus graded reinforcement learning reaches an overall score of 0.684, above 0.671 for binary reinforcement learning alone.

  5. arXiv

    Agent memory gets its own compressor: SumOne learns from masked reconstruction and cuts search-task cost to 22% of full context

    Long-horizon language-model agents accumulate ever-longer interaction histories, so context compaction is necessary, yet existing learned compaction methods often depend on downstream task outcomes or solver behavior. SumOne is a standalone 2-billion-parameter summarizer trained with a task-agnostic reward built from masked reconstruction and optimized with GRPO, using only raw trajectories and no reference summaries or downstream task execution. On DeepSearchQA it cuts cumulative agent input from 2.83M to 591K tokens with F1 0.48 versus 0.52 for full context; on SWE-bench Verified it reaches a 0.59 resolved rate versus 0.63, with input falling from 4.10M to 1.09M tokens.

  6. arXiv

    Long-horizon agents should bank credit per turn: MPO splits turn-level advantage from token refinement, lifting search pass@1 by up to 4.33 points

    Long-horizon LLM agents interact with their environment through complete turns but generate each action token by token, and existing reinforcement learning either gives a whole trajectory one shared advantage or treats every token as an action. MPO splits credit assignment into two scales: turn-level generalized advantage estimation carries credit across interactions, a mean-preserving local correction refines it within each turn, and the policy update uses complete-turn importance ratios. Across search, tool-use and automation tasks, MPO reports the highest score on all nine metrics at both Qwen3-4B-Thinking and Qwen3-30B-A3B-Instruct scales, including a 4.33-point BrowseComp-Plus pass@1 gain over the strongest non-MPO baseline at 4B.

  7. arXiv

    Chemistry figures no longer need manual transcription: RxnMiner lifts stereochemistry-aware reaction F1 from 21.0% to 71.0%

    Reaction figures in the chemical literature often pack several experiments into one shared template, substituent table and product grid, and recovering concrete reaction records from them has meant manual transcription. RxnMiner combines an expert-informed extraction procedure with a chemistry toolbox in one multimodal agent: it reads structures, binds evidence to each experimental entry, and instantiates undrawn reactants with a deterministic R-group engine. On ChemRxn-Bench, 603 reaction figures with 8,547 annotated records, it reaches 71.0% stereochemistry-aware reaction macro-F1 with a Qwen3.8-Flash-Next backbone, against 21.0% for direct prediction and 25.0% for the strongest existing extraction system.

  8. arXiv

    CANS gives every feed-forward neuron its own scaling coefficient, steering safety, toxicity and hallucination while holding general capability

    Behavior control in large language models often costs general capability; CANS instead assigns each feed-forward network neuron a scaling coefficient that accounts both for its contribution to the target behavior and for its disruption of general outputs. Target behavior is proxied by the log-probability of target tokens in model responses, capability preservation by output-distribution KL divergence on general-purpose text, and local approximations of the two yield a closed-form solution with no iterative training and no added inference-time computation. On Qwen3-4B and Llama models, CANS matches or beats SFT, CAA and neuron-level baselines on safety, toxicity and hallucination control while better preserving MMLU, BBH and IFEval performance.

  9. bioRxiv

    Folding models retooled as antibody screeners: frozen Protenix lifts Origin-1 macro AUROC from 0.669 to 0.757

    Antibody screening usually splits specificity, affinity and developability across separate models, and structural information is often compressed into a single confidence score such as ipTM. FoldAbS freezes the open-source folding model Protenix as a shared encoder, trains only lightweight task heads, and supplies specificity supervision with pseudo-negatives built from antigen mismatch, CDRH3 mutation and CDRH3 replacement. With Protenix as backbone, FoldAbS-Px raises macro AUROC on Origin-1 from 0.6690 to 0.7570 and reports the best results on affinity and four developability properties.

  10. arXiv

    Rooftop PV modules mapped without per-module labels: MV-PVMapper recovers 7,710 of 7,711 modules across three plants

    On five-direction UAV oblique surveys at roughly 1 cm nadir ground sampling distance, MV-PVMapper couples the frozen vision foundation model SAM 3 with photogrammetric geometry to output georeferenced polygons for individual modules without task-specific annotation or weight updates. Across three operational rooftop PV plants holding 7,711 reference modules, two feedback rounds match 7,710 with no false positives at an IoU threshold of 0.5; macro-averaged AJI reaches 97.55%, against 60.73% for an orthomosaic segmentation baseline and 41.24% for a multi-view pixel-voting baseline.