Your daily digest

Research papers, made simple to understand

The AI and machine learning papers people are reading most, each one rewritten in plain words: the problem, the idea, what was found and why it matters.

Older

Thursday, 12 November 2026

Newer
    • LLM Reasoning

    Structured textbook-based supervision improves mathematical reasoning in large language models

    TOME-Math converts 2,118 university-level math textbooks into 380K structured proof QA pairs, preserving curricular and dependency structure. Fine-tuning a 30B model on this corpus raises average benchmark scores by 16.7 points over the base and outperforms public baselines.

    Paper: Scaling Mathematical Proof Supervision from Structured Textbooks · Youqing Fang, Zhouqi Hua, Yinhao Tang and 7 others

    39 upvotes on Hugging Face6 min readRead the article

    • AI Agents
    • Training & Efficiency

    SteerDistill improves quantized LLM agents by teacher-guided correction and distillation

    SteerDistill uses a full-precision teacher to iteratively guide and correct a quantized LLM agent's failed actions, distilling these corrections for unassisted inference. It outperforms strong baselines on five benchmarks and reduces training cost by 67%.

    Paper: SteerDistill: Quantizing LLM Agents via Closed-Loop Teacher Steering · Genming Cui, Xiangyu Li, Zixu Hao and 3 others

    31 upvotes on Hugging Face7 min readRead the article

    • Reinforcement Learning & Alignment
    • Retrieval & RAG

    Reinforcement learning enables LLMs to self-learn from textbooks and improve problem solving

    This paper trains large language models to autonomously study and apply knowledge from structured textbooks using reinforcement learning, boosting mathematical proof accuracy from 51.5% to 79.5% and improving performance on other complex tasks.

    Paper: Incentivizing Self-Learning in Large Language Models via Reinforcement Learning · Lingkai Kong, Yuzhe Gu, Haiteng Zhao and 6 others

    27 upvotes on Hugging Face6 min readRead the article

Browse earlier papers

27 days of papers in October 2026

October 2026

Has papersYou are reading

  1. Chain-of-thought prompting improves reasoning in large language modelsand 2 more papers
  2. Intrinsic Distribution Matching improves one-step text-to-image generation with feature moment constraintsand 2 more papers
  3. Falcon enables efficient, real-time audio-visual target speaker extraction in complex scenesand 2 more papers
  4. Capability-driven recursive self-improvement evolves specialized multimodal agent modulesand 2 more papers
  5. Trajectory2Gym improves tool-use agents by synthesizing and grounding compositional interactionsand 2 more papers
  6. Survey unifies video understanding and generation via world modeling frameworkand 2 more papers
  7. Fine-tuning GPT-3 with human feedback improves instruction following and truthfulnessand 2 more papers
  8. Residual connections enable training of very deep neural networks with improved accuracyand 2 more papers
  9. WeVisOne-VL unifies object- and pixel-level vision-language tasks with structured token predictionand 2 more papers
  10. StreamEval benchmark measures interactive streaming video generation across five tasksand 2 more papers
  11. FocusGUI improves multi-step omni-modal GUI agents by dynamically prioritizing relevant inputsand 2 more papers
  12. Parameter-efficient fine-tuning scales large multimodal models for brain encodingand 2 more papers
  13. HexaShell uses coding agents with RGB triangulation to improve robot manipulation on RoboDojoand 2 more papers
  14. Two efficient models learn high-quality word vectors from huge text corporaand 2 more papers
  15. Transformer replaces recurrence with multi-head self-attention for faster, better translationand 2 more papers
  16. Waypoint-OPD improves RL policy transfer by gating shifts using intermediate checkpointsand 2 more papers
  17. ATAS framework improves spatial and temporal consistency in agentic video generationand 2 more papers
  18. Jeva architecture enables stable, scalable, and cacheable system one decisionsand 2 more papers
  19. GroundingPI improves precise visual grounding for physical intelligence tasksand 2 more papers
  20. DiffusionART stabilizes and speeds up reinforcement learning fine-tuning of diffusion modelsand 2 more papers
  21. LoGRA cuts LLM RL training memory up to 45.7% using low-rank gradient sketchesand 2 more papers
  22. UltraText Bench evaluates dense bilingual visual text rendering in image generationand 2 more papers
  23. AgentGarten enables evolving agents via code-driven worlds and neural renderingand 2 more papers
  24. STEPQuant reduces memory and preserves accuracy in low-bit quantization of recurrent statesand 2 more papers
  25. Compact strict alignment supervision outperforms broader coverage in cross-tokenizer distillationand 2 more papers
  26. Kandinsky 6.0 Video generates synchronized 5-second videos with 44 kHz audio using dual-stream diffusion modelsand 2 more papers
  27. RealCompanion benchmarks AI understanding of real long-term human conversationsand 2 more papers