AI Dataset2026
Provides 10,000 agentic multi-turn coding and reasoning traces from Fable 5.1 with step-by-step chain-of-thought and tool-use, heavily deduplicated and filtered; totals ~500M tokens (1.97 GB). Suited for SFT, distillation, and training long-horizon reasoning and code-generation models.
AI Agent Papers2026
Jianlyu Chen, Yuyang Hu +9Beijing Academy of Artificial Intelligence, University of Science and Technology of China +2
Distills operational know‑how from ML GitHub repositories into compact, verified 'skills' that research agents can load and reuse. Produces a skill format (SKILL.md, references, scripts), the AREX‑Skill Library (5,000+ skills from 1,000 repos), and demonstrates sizable benchmark gains when agents use skills.
Selectively admits dense token-level teacher supervision only after a prompt-level verifier audit, routing prompts that fail the audit to verifier-grounded trajectory supervision instead — reducing harmful updates from confidently wrong teachers and improving teacher GPU utilization.
AI Dataset2026
Contains 5,000 coding and chain-of-thought reasoning traces generated by Fable 5.1 — ~150M tokens of step-by-step programming CoT. Deduplicated and filtered for high quality; intended for supervised fine-tuning and distillation to improve reasoning in smaller models.
AI Dataset2026
A curated set of 1,000 high-quality chain-of-thought coding and reasoning traces generated by Fable 5.1, totaling ~30M tokens (109 MB). Designed for SFT/distillation to teach smaller models step-by-step programmatic reasoning and debugging.
AI Video Papers2026
Provides an end-to-end, reproducible foundation for camera‑controllable, long‑horizon video world models — converting 1.43M clips from 10 datasets into a unified canonical corpus and releasing data, pipelines, recipes, and weights. Introduces backbone‑native adaptation and a three‑stage training recipe to produce 5B–33B models that enable minute‑to‑hour rollouts after training on 5s sequences.
Uses lightweight discrete-diffusion adapters to draft blocks of tokens in parallel while keeping the original autoregressive model as the final arbiter, enabling lossless speedups (up to 3×) and compatibility with existing AR weights via a Diffusion Distillation phase.
Zixuan Fu, Bingxiang He +11Affiliation: University of Chinese Academy of Sciences, Affiliation: Tsinghua University +3
Studies on-policy distillation (OPD) at the data-minimal limit by training on a single query, measuring state coverage and alignment dynamics, and showing OPD is often data-overfed but algorithm-starved.
Converts a natural-language function specification into a reusable local neural function by using teacher models to synthesize examples and finetuning a small adapter for a compact interpreter. Achieves higher semantic accuracy (83.6% on FuzzyBench-Hard) at the cost of roughly one minute compile time; produces versionable PAW artifacts for local deployment.
Builds high-fidelity image generators by pairing a 6B Diffusion Transformer with a frozen LLaDA2.0-Mini vision-language module, relying on extensive image-only pretraining and mid-training; model weights, training code, and recipes are released.
Trains end-to-end driving without human trajectory supervision by decoupling perception and action: DriveVFM distills multiple frozen vision foundation models into a single camera backbone, and DriveRL trains a privileged closed-loop RL teacher whose rollouts supervise a camera-only planner, yielding state-of-the-art closed-loop benchmark results.
Investigates long-horizon memorization in language models by evaluating how complementary continual-learning mechanisms (data, function, weight anchors and low-rank allocation/LoRA) compose to reduce catastrophic forgetting across 100 sequential tasks; best composition raises final retention from 1.2% to 34.9%.