Daily Papers of 2026-02-17

  1. Experiential Reinforcement Learning 66 upvotes, #1 of 2026-02-17
  2. DeepImageSearch: Benchmarking Multimodal Agents for Context-Aware Image Retrieval in Visual Histories 51 upvotes, #2 of 2026-02-17
  3. BitDance: Scaling Autoregressive Generative Models with Binary Tokens 48 upvotes, #3 of 2026-02-17
  4. Nanbeige4.1-3B: A Small General Model that Reasons, Aligns, and Acts 29 upvotes, #4 of 2026-02-17
  5. REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents 23 upvotes, #5 of 2026-02-17
  6. STATe-of-Thoughts: Structured Action Templates for Tree-of-Thoughts 20 upvotes, #6 of 2026-02-17
  7. Query as Anchor: Scenario-Adaptive User Representation via Large Language Model 18 upvotes, #7 of 2026-02-17
  8. InnoEval: On Research Idea Evaluation as a Knowledge-Grounded, Multi-Perspective Reasoning Problem 16 upvotes, #8 of 2026-02-17
  9. Data Darwinism Part I: Unlocking the Value of Scientific Data for Pre-training 15 upvotes, #9 of 2026-02-17
  10. Learning to Configure Agentic AI Systems 14 upvotes, #10 of 2026-02-17
  11. Qute: Towards Quantum-Native Database 13 upvotes, #11 of 2026-02-17
  12. VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction 12 upvotes, #12 of 2026-02-17
  13. UniWeTok: An Unified Binary Tokenizer with Codebook Size 2^{128} for Unified Multimodal Large Language Model 12 upvotes, #12 of 2026-02-17
  14. Embed-RL: Reinforcement Learning for Reasoning-Driven Multimodal Embeddings 9 upvotes, #14 of 2026-02-17
  15. BrowseComp-V^3: A Visual, Vertical, and Verifiable Benchmark for Multimodal Browsing Agents 8 upvotes, #15 of 2026-02-17
  16. WebWorld: A Large-Scale World Model for Web Agent Training 7 upvotes, #16 of 2026-02-17
  17. AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories 6 upvotes, #17 of 2026-02-17
  18. LaViDa-R1: Advancing Reasoning for Unified Multimodal Diffusion Language Models 5 upvotes, #18 of 2026-02-17
  19. Conversational Image Segmentation: Grounding Abstract Concepts with Scalable Supervision 4 upvotes, #19 of 2026-02-17
  20. FireRed-Image-Edit-1.0 Techinical Report 4 upvotes, #19 of 2026-02-17
  21. AIDev: Studying AI Coding Agents on GitHub 3 upvotes, #21 of 2026-02-17
  22. MoRL: Reinforced Reasoning for Unified Motion Understanding and Generation 3 upvotes, #21 of 2026-02-17
  23. CellMaster: Collaborative Cell Type Annotation in Single-Cell Analysis 2 upvotes, #23 of 2026-02-17
  24. LM-Lexicon: Improving Definition Modeling via Harmonizing Semantic Experts 2 upvotes, #23 of 2026-02-17
  25. Blind to the Human Touch: Overlap Bias in LLM-Based Summary Evaluation 1 upvotes, #25 of 2026-02-17
  26. Benchmarking Knowledge-Extraction Attack and Defense on Retrieval-Augmented Generation 1 upvotes, #25 of 2026-02-17
  27. DHPLT: large-scale multilingual diachronic corpora and word representations for semantic change modelling 1 upvotes, #25 of 2026-02-17
  28. Acoustivision Pro: An Open-Source Interactive Platform for Room Impulse Response Analysis and Acoustic Characterization 1 upvotes, #25 of 2026-02-17
  29. Can I Have Your Order? Monte-Carlo Tree Search for Slot Filling Ordering in Diffusion Language Models 1 upvotes, #25 of 2026-02-17
  30. Preliminary sonification of ENSO using traditional Javanese gamelan scales 1 upvotes, #25 of 2026-02-17
  31. Exposing the Systematic Vulnerability of Open-Weight Models to Prefill Attacks 1 upvotes, #25 of 2026-02-17
  32. EditCtrl: Disentangled Local and Global Control for Real-Time Generative Video Editing 1 upvotes, #25 of 2026-02-17
  33. Found-RL: foundation model-enhanced reinforcement learning for autonomous driving 0 upvotes, #33 of 2026-02-17
  34. SPILLage: Agentic Oversharing on the Web 0 upvotes, #33 of 2026-02-17
  35. A Critical Look at Targeted Instruction Selection: Disentangling What Matters (and What Doesn't) 0 upvotes, #33 of 2026-02-17
  36. Knowing Isn't Understanding: Re-grounding Generative Proactivity with Epistemic and Behavioral Insight 0 upvotes, #33 of 2026-02-17

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.