Evaluates a weak teacher's RL-induced policy shift on the student's own rollouts and amplifies verifier-supported updates so stronger models can learn from weaker supervisors and surpass them. It rescales only verifier-supported policy-gradient components to preserve optimization fixed points while accelerating learning, reducing student updates versus standard RL or distillation.
AI Agent Papers2026
Explores a practical mechanism for recursive self-improvement by post-training LLMs: uses a routing harness to record agent executions and convert traces into curriculum-guided supervised fine-tuning and on-policy distillation data, closing an evaluation-selection-update loop and improving benchmark performance.
An 8B native unified multimodal model that understands, reasons about, and generates high-fidelity visual content without a separate visual encoder or VAE—focused on spatially coherent patch reconstruction, native 4K generation, multi-expert distillation, and improved text rendering and image editing.
Haojun Zhang, Yi Zou +8Affiliation: Xini Ding, Hao Li, Shuchang Zhou, Xianming Liu, Shiyu Huang‡, Affiliation: XPeng Inc.
Progressively prunes and distills audio encoders for speech LLMs to cut inference cost while preserving decoder-facing embeddings, using behavioral probes, representation alignment, cross-scale distillation and LoRA finetuning; reports reduced macro-error on Chinese–English benchmarks.
Analyzes why on-policy distillation can produce excessively long student rollouts by identifying termination-token mismatch between students and teachers, and proposes aggregating functionally equivalent EOS tokens as a shared semantic stopping action to mitigate length inflation.
Teaches vision-language models to predict and integrate physical-world state transitions from interaction trajectories (observation → action → next observation). Introduces a three-level curriculum and the LSI-108K dataset, and applies supervised fine-tuning plus on-policy distillation to improve local transition modeling and long-horizon spatial integration.
AI Video Papers2026
Wangbo Yu, Kunhao Liu +9Affiliation: Peking Universityhttps://drexubery.github.io/WorldCrafter, Affiliation: ARC Lab, Tencent IEG
Learns a camera-queryable implicit 3D-aware memory that compresses multi-view history into target-view tokens to enable long-horizon, camera-controllable video generation. Improves revisit consistency and camera-control accuracy and supports streaming exploration from a single image or text prompt.
AI Model2026
A 9B agentic multimodal SFT checkpoint distilled from Qwen3.5-9B for coding, general agent tasks, visual coding and cybersecurity. Provided by Xiaomi MiMo as a research seed (77.4B-token SFT mix) to bootstrap agentic RL and tool-use experiments.
AI Image2026
A distilled LoRA adapter for Qwen-Image-2.1 that runs text-to-image generation and instruction-driven image editing in a few transformer passes (shipped as a 6-step r256 LoRA). Samples with a fixed sigma schedule, no classifier-free guidance; non-commercial research license.
Shows that private post-training changes leave measurable "behavioral shadows" in task-unrelated single-word outputs and proposes Active Taskless Distillation (ATD) to transfer capabilities to a student using only those single-word teacher responses, without teacher logits or parameters.
Xingyu Miao, Zizun Li +46Shanghai AI Laboratory: Affiliation: Project page:https://internrobotics.github.io/InternW0-Delta/
Learns joint predictive visual dynamics and action generation for generalist robot manipulation, translating future-relevant visual representations into actions. Integrates a Mixture-of-Transformers coupling a video expert and action expert, a frozen vision-language model for semantics, 4D distillation, and Causal Imprint; pretrained on a 20K+ hour heterogeneous corpus.
Julianna Piskorz, Antonin Berthon +1Affiliation: University of Cambridge, Affiliation: Cambridge, UK
Investigates how rollout policy, token-level KL direction, and learning rate each affect LLM distillation across Llama3 and Qwen2.5 on reasoning tasks; finds KL direction and learning rate dominate outcomes while rollout policy has a modest effect.