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。
Xiyuan Yang, Sheikh Sarwar, Jingru Cheng +8 authors
World Model RL replaces costly environment execution with a learned world model and applies debiasing and denoising to accelerate post-training of autonomous research agents.
50 篇论文 · 按点赞排序
NeoHorse Team, Guoliang Cao, Guohao Dai +34 authors
NeoHorse-1 uses agentic post-training with intelligent routing, structured feedback loops, and curriculum-based distillation to improve model capabilities across agent benchmarks.
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.
NCP Team, Jiaqi Cao, Chiyu Chen +25 authors
NCP-ArchPreview is a large latent-space language model that jointly trains next-token and next-concept prediction to improve pretraining efficiency and downstream performance.
Dingjie Song, Hanrong Zhang, Dawei Liu +10 authors
Command-line coding agents (e.g., Claude Code, Gemini CLI) can already read and write files and sustain long sessions, yet end-to-end research still fragments across chat tools, IDEs, terminals, and writing environments, and the decisions that make it auditable are rarely preserved. We present Dr. Claw, an open-source workspace that wraps existing coding-agent executors in a controllable and auditable human-in-the-loop workflow rather than introducing another autonomous agent. Persistent state objects, a reusable skill library, and multi-executor coordination link human decisions to AI execution, turning planning, execution, and writing into one traceable, recoverable loop. We demonstrate Dr. Claw through an interactive three-view scenario and a failure-recovery walkthrough, and evaluate it against a bare command-line agent sharing the same backend executor, so the comparison contrasts the whole orchestration layer (task graph, state objects, and skill library) with the agent it wraps. Holding the executor fixed, Dr. Claw scores higher on research completeness while persisting an auditable, recoverable process trail. Demo access: repository https://github.com/OpenLAIR/dr-claw, released under AGPL-3.0 with GPL-3.0 upstream components.
Haiwen Diao, Jiahao Wang, Chenjing Ding +62 authors
SenseNova-U1.5 is an 8B native unified multimodal model that performs visual understanding, reasoning, and generation without encoders or VAEs, achieving high fidelity and instruction following through patch reconstruction, curated data, expert optimization, and on-policy distillation.
Yihang Chen, Yuxiang Chen, Yuxuan Huang +3 authors
The study formalizes multi-agent LLM coordination via bilevel games and stochastic memory reflection, introducing a grounded evaluation gate and SRMA algorithm with convergence guarantees, validated on SWE-bench.
Subham Sekhar Sahoo, Lingjie Chen, Khiem Pham +14 authors
Diffusion-augmented autoregressive language models use parallel token sampling via distilled diffusion weights and a specialized sampler to accelerate inference without quality loss or draft models.
Yanzhe Chen, Zechen Bai, Zhijun Cao +7 authors
Show-Harness links vision-language models to robot control via discrete semantic actions interpreted by embodiment-specific modules, enabling zero-shot and efficient fine-tuned deployment across robots and GUIs.
Orantqing, Shengpeng Ji, Junlong Tong +20 authors
Gander is an end-to-end framework that integrates continuous multi-modal streaming, real-time full-duplex interaction, and agentic reasoning through a Cerebellum-Brain architecture and a chunk-level token stream design.
Zixun Huang, Kishan Panaganti, Haitao Mi +1 authors
FlowBalance improves reasoning models via verifier-calibrated self-guidance using trajectory-level score reweighting and profile-based trajectory balance.
Zheng-Hui Huang, Guixu Lin, Jiacheng Lin +8 authors
A programmable world model separates explicit state evolution from video generation using executable rules and 3D bounding boxes to maintain persistent, controllable environments.
Jaewon Chu, Jinwoo Seo, Jaewon Cho +4 authors
AgentGrad improves multi-agent prompt optimization by identifying target agents through sequential intervention and clustering gradients semantically to avoid mixing unrelated errors.
Youngrok Park, Sangmin Bae, Hojung Jung +6 authors
On-Policy Reverse Distillation enables stronger models to exceed weak supervisors by amplifying verifier-supported policy gradients along the teacher's shift direction, accelerating optimization without imposing capacity limits.
Soohyun Ryu, Sohee Kim, Eunho Yang
Large vision-language models trained on synthetic block-manipulation tasks improve 3D spatial reasoning and generalize to real-world visual tasks.
Yuran Wang, Siqiao Huang, Mingleyang Li +21 authors
OpenWAM factorizes world-action pretraining into modular components to identify key design principles, yielding a scalable open model with strong simulation and real-robot performance.
Ziyuan Liu, Hengqi Liu, Zichuan Wang +6 authors
Two large-scale search agents are trained via a multi-stage pipeline combining supervised fine-tuning and reinforcement learning against live search, achieving state-of-the-art open-source results on complex web benchmarks through rigorous trajectory filtering and inference-time context management.
Igor Pavlovic, Thiemo Wandel, Anton Obukhov +6 authors
Marigold V2 repurposes diffusion transformers for monocular depth estimation via single-step flow-matching inference, semantic alignment, and a Sinkhorn-based two-stage fine-tuning protocol, yielding sharper out-of-distribution depth maps and strong results on related dense regression tasks.
Hao He, Chengcheng Hu, Zirun Su +17 authors
DriveZero is an end-to-end autonomous driving system that combines a vision foundation model for perception with a closed-loop reinforcement learning action model to learn driving behaviors beyond human demonstrations.
RadixArk, Tom Chen, Mao Cheng +10 authors
Miles is an open-source, production-ready system for large-scale reinforcement learning and post-training that supports diverse backends, weight synchronization, LoRA, distillation, and diffusion models.
AgiBot Research Team, Renhang Liu, Wenzhi Zhao +42 authors
GE-Act 2.0 is a world-action model trained from scratch with a control-oriented autoencoder, single-step visual planner, and inverse dynamics model, using knowledge-aligned selective optimization to enable scalable zero-shot robot manipulation across diverse skills and conditions.
Junyao Yang, Yucheng Shi, Zhongzhi Li +4 authors
T1 is a 122B Mixture-of-Experts model trained with reinforcement learning to execute long-horizon terminal tasks in a cloud sandbox, achieving state-of-the-art results through stable actor-critic optimization and out-of-distribution training.
Nanxi Li, Yingzi Ma, Yulong Cao +4 authors
EvoSafeHarness optimizes deployable safety harnesses by jointly searching natural-language policies and executable logic tailored to a frozen model and target domain, improving safety-utility trade-offs across agent benchmarks.
Zhuoran Zhao, Shengju Qian, Tongtong Liang +7 authors
Mask Forcing mitigates mode collapse in distilled autoregressive video diffusion by injecting masked cleaner signals during self-rollout, improving visual quality without extra training data.
Xingjian Ran, Xiaoye Mo, Sihao Liu +3 authors
SceneMosaic combines learned image priors with vision-language agents to efficiently generate diverse, physically valid indoor scenes by evolving local units and composing them globally.
Javier del Pino, Salvador Rodríguez, Alejandro Garabito +2 authors
ENEAS unifies text-prompted instance tracking and open-concept semantic discovery via temporal memory extension and a verification layer that filters visual distractors.
Chengqian Ma, Wei Tao, Haoyu Zhang +1 authors
Motion-Omni is an end-to-end framework that jointly generates spoken dialogue and full-body co-speech motion from shared hidden states, using scalable pseudo-labeling and a unified evaluation protocol to achieve real-time, aligned responses.
Sagi Polaczek, Noa Kraicer, Gal Metzer +4 authors
Audio-video diffusion models exhibit bidirectional semantic leakage through cross-modal attention pathways, which can be diagnosed via attention-derived signals and mitigated through inference-time alignment interventions.
Janghyeon Kim, Minsoo Kim, Kyuhong Shim +1 authors
BeaconKV improves memory efficiency for long reasoning traces by using compact beacon queries to predict which past key-value pairs will be revisited, reducing cache size without sacrificing accuracy.
Yuxing Lu, Yicheng Chen, Shanchan Wu +1 authors
A procedural graph framework organizes agent actions into structured relational triplets, providing situational guidance and self-evolving topology to improve long-horizon tool use.
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