Submitted by yxsong 129 Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference · 22 authors 17
Submitted by OrlandoHugBot 59 Skywork UniPic: Unified Autoregressive Modeling for Visual Understanding and Generation · 15 authors 757 2
Submitted by Jianxiong 50 LongVie: Multimodal-Guided Controllable Ultra-Long Video Generation · 8 authors 82 3
Submitted by zsytony 33 CompassVerifier: A Unified and Robust Verifier for LLMs Evaluation and Outcome Reward · 11 authors 43 4
Submitted by jiawei-ucas 17 LiveMCPBench: Can Agents Navigate an Ocean of MCP Tools? · 9 authors 51 5
Submitted by SanghyeokLee 15 Representation Shift: Unifying Token Compression with FlashAttention · 6 authors 19 2
Submitted by xxiaoyali 13 CRINN: Contrastive Reinforcement Learning for Approximate Nearest Neighbor Search · 5 authors 2
Submitted by Bohan22 11 Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction · 20 authors 3
Submitted by Jinsong-Li 8 LAMIC: Layout-Aware Multi-Image Composition via Scalability of Multimodal Diffusion Transformer · 6 authors 18 2
Submitted by ahnpersie 5 ChartCap: Mitigating Hallucination of Dense Chart Captioning · 3 authors 1 2
Submitted by Liang-ZX 5 HyCodePolicy: Hybrid Language Controllers for Multimodal Monitoring and Decision in Embodied Agents · 10 authors 2
Submitted by AmineAllo 4 What Is Your AI Agent Buying? Evaluation, Implications and Emerging Questions for Agentic E-Commerce · 5 authors 5 2
Submitted by chaitanya100100 4 UniEgoMotion: A Unified Model for Egocentric Motion Reconstruction, Forecasting, and Generation · 6 authors 21 2
Submitted by DanCip 3 TreeRanker: Fast and Model-agnostic Ranking System for Code Suggestions in IDEs · 4 authors 2
Submitted by ikodoh 3 Bidirectional Likelihood Estimation with Multi-Modal Large Language Models for Text-Video Retrieval · 5 authors 13 2
Submitted by SecureLLMSys 2 AttnTrace: Attention-based Context Traceback for Long-Context LLMs · 4 authors 3 2
Submitted by amanchadha 2 AlignGuard-LoRA: Alignment-Preserving Fine-Tuning via Fisher-Guided Decomposition and Riemannian-Geodesic Collision Regularization · 4 authors 2
Submitted by amanchadha 1 TRACEALIGN -- Tracing the Drift: Attributing Alignment Failures to Training-Time Belief Sources in LLMs · 3 authors 2