Explore · 每周精选
发现最受关注的研究论文,追踪研究趋势,订阅感兴趣的期刊与关键词。
875 篇论文 · 按点赞排序
Yafu Li, Runzhe Zhan, Haoran Zhang +25 authors
A systematic approach transforms post-trained reasoning models into rigorous olympiad-level solvers through reverse-perplexity curriculum, two-stage reinforcement learning, and test-time scaling, achieving gold-medal performance on mathematical and physics competitions.
Wenbo Pan, Zhichao Liu, Shujie Liu +6 authors
Taste-Bench evaluates agent decision-making at trajectory forks to measure long-horizon judgment, revealing limited frontier model accuracy and demonstrating trainable improvements via distillation.
Shun Zou, Ziyu Ma, Yi Zou +6 authors
TerminalHorizon automates long-horizon terminal agent data synthesis by reconstructing environments around omitted work and progressively expanding tasks, yielding substantial gains across coding, production, and reasoning benchmarks.
Hanchen Xia, Baoyou Chen, Yutang Ge +5 authors
REMORY enhances long-horizon agent memory by adding learnable soft tokens to textual summaries, improving decision accuracy with minimal context usage.
Zhifei Xie, Jiaqi Lang, Ze An +7 authors
VoiceMem introduces a dual-brain streaming memory architecture for speech language models that improves retrieval accuracy, emotional personalization, and real-time efficiency.
Xiaofei Wen, Wenjie Jacky Mo, Xingyu Fu +6 authors
Video-capable multimodal large language models exhibit apparent audio understanding driven by visual cues rather than actual audio processing, necessitating intervention-based frameworks for diagnosing and improving audio-visual alignment.
Keyu Tu, Zhuowei Chen, Mengqi Huang +4 authors
Semantic task completion video generation evaluates whether generated videos achieve intended outcomes with semantic grounding, supported by a curated dataset and vision-language model-based benchmark.
Wangbo Yu, Kunhao Liu, Wenbo Hu +8 authors
WorldCrafter uses a camera-queryable implicit 3D-aware memory to maintain long-horizon consistency and camera control in video world models during extended scene exploration.
Ziyang Ma, Zhikang Niu, Wenming Tu +30 authors
AuK is an open-source foundational model that unifies speech generation and editing via natural-language instructions and audio context, using a multimodal language model, joint VAE, hybrid rectified-flow Transformer, and efficient distillation for fast inference.
Yu Zhang, Ruiqi Li, Changhao Pan +3 authors
SwanTale is a multi-speaker expressive speech and audio generation model that supports both instruction-based and zero-shot synthesis with high-quality multi-modality outputs.
SII-OpenMOSS Team, Donghua Yu, Mingshu Chen +37 authors
MOVA is an open-source model that generates synchronized audio-visual content using a Mixture-of-Experts architecture with 32 billion parameters, supporting image-text to video-audio generation tasks.
Meng Luo, Yanlin Li, Hao Li +7 authors
AI applications across the game lifecycle are organized into six roles, highlighting transferable capabilities and setting-specific limitations that require targeted validation.
Simple AI, Yuteng Wei, Jinming Ma +15 authors
High-fidelity robot-free UMI data enables direct real-robot deployment without teleoperation anchors, improving manipulation policy accuracy and success rates.
Veuns-Team, Changlong Gao, Zhangxuan Gu +24 authors
UI-Venus-1.5 is a unified GUI agent with improved performance through mid-training stages, online reinforcement learning, and model merging techniques.
Yinjie Wang, Xuyang Chen, Xiaolong Jin +2 authors
OpenClaw-RL framework enables policy learning from diverse next-state signals across multiple interaction modalities using asynchronous training with PRM judges and hindsight-guided distillation.
Jianyu Wu, Yizhou Wang, Encheng Su +2 authors
Dual-Anchored Policy Distillation resolves privilege illusion in on-policy self-distillation by aligning matched-information paths and reducing reliance on privileged guidance, improving performance across model scales.
Fengkai Yang, Zherui Chen, Xiaohan Wang +10 authors
Group-based reinforcement learning from verifier rewards suffers from biased advantage estimation that underestimates hard prompts and overestimates easy prompts, which is addressed through a history-aware adaptive difficulty weighting method that improves performance on mathematical reasoning benchmarks.
Kaijin Chen, Dingkang Liang, Xin Zhou +4 authors
Hybrid Memory enables video world models to maintain consistent tracking of dynamic subjects during occlusion by combining archival storage for static backgrounds with active tracking for moving objects, using a specialized architecture with tokenized memory and spatiotemporal retrieval mechanisms.
Yibo Li, Jinhang Qiu, Zhi Zheng +4 authors
Learn2Play Bench evaluates LLM agents on novel text-based games to measure learning from interaction rather than pretrained reasoning, revealing benefits of full experience retention, a human performance gap, and harness design effects.
Yawen Luo, Xiaoyu Shi, Junhao Zhuang +5 authors
ShotStream enables real-time interactive multi-shot video generation through causal architecture design, dual-cache memory mechanisms, and two-stage distillation to maintain visual consistency and reduce latency.
Wenxuan Huang, Yu Zeng, Qiuchen Wang +12 authors
Vision-DeepResearch introduces a multimodal deep-research paradigm enabling multi-turn, multi-entity, and multi-scale visual and textual search with deep-research capabilities integrated through cold-start supervision and reinforcement learning.
Daxiang Dong, Mingming Zheng, Dong Xu +17 authors
Qianfan-OCR is a 4B-parameter vision-language model that unifies document parsing, layout analysis, and understanding while maintaining strong performance across multiple OCR benchmarks through its Layout-as-Thought mechanism.
Junyoung Seo, Hyunwook Choi, Minkyung Kwon +10 authors
What if a world simulation model could render not an imagined environment but a city that actually exists? Prior generative world models synthesize visually plausible yet artificial environments by imagining all content. We present Seoul World Model (SWM), a city-scale world model grounded in the real city of Seoul. SWM anchors autoregressive video generation through retrieval-augmented conditioning on nearby street-view images. However, this design introduces several challenges, including temporal misalignment between retrieved references and the dynamic target scene, limited trajectory diversity and data sparsity from vehicle-mounted captures at sparse intervals. We address these challenges through cross-temporal pairing, a large-scale synthetic dataset enabling diverse camera trajectories, and a view interpolation pipeline that synthesizes coherent training videos from sparse street-view images. We further introduce a Virtual Lookahead Sink to stabilize long-horizon generation by continuously re-grounding each chunk to a retrieved image at a future location. We evaluate SWM against recent video world models across three cities: Seoul, Busan, and Ann Arbor. SWM outperforms existing methods in generating spatially faithful, temporally consistent, long-horizon videos grounded in actual urban environments over trajectories reaching hundreds of meters, while supporting diverse camera movements and text-prompted scenario variations.
Yu Wang, Yi Wang, Rui Dai +4 authors
Urban socio-semantic segmentation is achieved through a vision-language model framework that combines cross-modal recognition and multi-stage reasoning with reinforcement learning optimization.
Yifei Wang, Yueqi Wang, Zhenrui Yue +6 authors
FASA is a novel framework that uses query-aware token eviction and functional sparsity in RoPE to reduce KV cache memory usage while maintaining high performance in long-context LLM tasks.
Shuqi Lu, Chaofan Li, Kun Luo +21 authors
AREX is a recursively self-improving deep research agent that verifies answers constraint-wise, compresses verified evidence into a compact state, and uses targeted follow-up research to refine results over long horizons.
Haolin He, Yunfei Chu, Qi Chen +15 authors
OmniVChat-Studio synthesizes audio-visual dialogues to train and evaluate omni models via reinforcement learning, improving real-world dialogue performance.
Zhanhui Zhou, Lingjie Chen, Hanghang Tong +1 authors
A unified open-source framework is presented that standardizes core components of diffusion language modeling for reproduction, customization, and accessible development of both large and small models.
Heng Wang, Jielin Qiu, Wenting Zhao +7 authors
Random eviction of reasoning tokens matches selective KV cache compression because reasoning traces are self-protecting through redundancy, making scoring unnecessary once prompts are preserved.
Xunhao Lai, Weiqi Xu, Yufeng Yang +8 authors
MiniMax Sparse Attention enables efficient processing of ultra-long contexts in large language models through blockwise sparsity and optimized GPU execution, achieving significant speedups while maintaining performance.
北京市昌平区探索星信息技术及软件开发工作室
京ICP备2026059466号