Daily Papers of 2026-07-07

  1. OmniOpt: Taxonomy, Geometry, and Benchmarking of Modern Optimizers 75 upvotes, #1 of 2026-07-07
  2. UI-MOPD: Multi-Platform On-Policy Distillation for Continual GUI Agent Learning 70 upvotes, #2 of 2026-07-07
  3. PixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space 63 upvotes, #3 of 2026-07-07
  4. ResearchStudio-Reel: Automate the Last Mile of Research from Paper to Poster, Video, and Blog 61 upvotes, #4 of 2026-07-07
  5. ResearchStudio-Idea: An Evidence-Grounded Research-Ideation Skill Suite from ML Conference Outcomes 54 upvotes, #5 of 2026-07-07
  6. MANCE: Manifold Aware Concept Erasure 46 upvotes, #6 of 2026-07-07
  7. Vision Pretraining for Dense Spatial Perception 43 upvotes, #7 of 2026-07-07
  8. GigaWorld-1: A Roadmap to Build World Models for Robot Policy Evaluation 37 upvotes, #8 of 2026-07-07
  9. Bridging Interleaved Multi-Modal Reasoning as a Unified Decision Process 37 upvotes, #8 of 2026-07-07
  10. Wan-Streamer v0.2: Higher Resolution, Same Latency 37 upvotes, #8 of 2026-07-07
  11. Multi-Turn Agentic Scientific Literature Search via Workflow Induction 28 upvotes, #11 of 2026-07-07
  12. EVA-Client: A Unified Data Collection, Inference, and Deployment Framework for Embodied Policies on Real Robots 26 upvotes, #12 of 2026-07-07
  13. InternVLA-A1.5: Unifying Understanding, Latent Foresight, and Action for Compositional Generalization 25 upvotes, #13 of 2026-07-07
  14. Multiplayer Interactive World Models with Representation Autoencoders 24 upvotes, #14 of 2026-07-07
  15. Do All Visual Tokens Matter Equally? Object-Evidence Preserving Token Merging for Vision-Language Retrieval 22 upvotes, #15 of 2026-07-07
  16. ACID: Action Consistency via Inverse Dynamics for Planning with World Models 21 upvotes, #16 of 2026-07-07
  17. KVpop -- Key-Value Cache Compression with Predictive Online Pruning 21 upvotes, #16 of 2026-07-07
  18. Unified Audio Intelligence Without Regressing on Text Intelligence 20 upvotes, #18 of 2026-07-07
  19. Look Before You Leap: Distilling Tree Search into Action Evaluation for Frozen VLA Models 19 upvotes, #19 of 2026-07-07
  20. Perceptual Flow Matching for Few-Step Generative Modeling 17 upvotes, #20 of 2026-07-07
  21. dOPSD: On-Policy Self-Distillation for Diffusion Language Models 17 upvotes, #20 of 2026-07-07
  22. EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments 17 upvotes, #20 of 2026-07-07
  23. LLM-as-a-Verifier: A General-Purpose Verification Framework 14 upvotes, #23 of 2026-07-07
  24. SeKV: Resolution-Adaptive KV Cache with Hierarchical Semantic Memory for Long-Context LLM Inference 11 upvotes, #24 of 2026-07-07
  25. MV-Forcing: Long Multi-View Video Generation via 4D-Grounded Spatio-Temporal Self-Forcing 10 upvotes, #25 of 2026-07-07
  26. Deform360: A Massive Multi-view Visuotactile Dataset for Deformable World Models 10 upvotes, #25 of 2026-07-07
  27. PraMem: Practice-derived Experiential Memory for Long-horizon Behavior Prediction 9 upvotes, #27 of 2026-07-07
  28. Mastermind: Strategy-grounded Learning for Repository-Scale Vulnerability Reproduction 8 upvotes, #28 of 2026-07-07
  29. GORGO: Online Tuning for Cross-Region Network-Aware LLM Serving 7 upvotes, #29 of 2026-07-07
  30. Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification 7 upvotes, #29 of 2026-07-07
  31. AI Wizards at EXIST 2026: Hierarchical Soft-Label Learning for Multimodal Sexism Identification in Memes 7 upvotes, #29 of 2026-07-07
  32. Transition-Aware best-of-N sampling for Longitudinal Chest X-ray Reports 6 upvotes, #32 of 2026-07-07
  33. CONFLUX: A Latent Diusion Model for 3D Chest-CT Synthesis with RL Post-Training 6 upvotes, #32 of 2026-07-07
  34. Learning to Trigger: Reinforcement Learning at the Large Hadron Collider 5 upvotes, #34 of 2026-07-07
  35. GaP: A Graph-as-Policy Multi-Agent Self-Learning Harness For Variational Automation Tasks 5 upvotes, #34 of 2026-07-07
  36. Speaker-Aware Temporal Aggregation Strategies on Segment Representations for Depression Detection in Dyadic Interaction: A Benchmark Study 4 upvotes, #36 of 2026-07-07
  37. Taste-aware music retrieval from audio embeddings 4 upvotes, #36 of 2026-07-07
  38. SynCity 3000: Bootstrapping Scene-Scale 3D Diffusion 4 upvotes, #36 of 2026-07-07
  39. PixCon: Clean-Positive Contrastive Learning for Foundation-Model Semi-Supervised Segmentation 3 upvotes, #39 of 2026-07-07
  40. Speaker-Disentangled Chunk-Wise Regression for Syllabic Tokenization 3 upvotes, #39 of 2026-07-07

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.