AIAIAny
  • Search
  • Collection
  • Category
  • Tag
  • Daily AI
AIAIAny

Category

Explore by categories

AIAIAny

Curated AI Resources for Everyone

[email protected]

Powered by airss.app

Product
  • Search
  • Collection
  • Category
  • Tag
Resources
  • Blog
Company
  • Privacy Policy
  • Terms of Service
  • Sitemap
Copyright © 2026 All Rights Reserved.
  • All Categories

  • AI Leaderboard

  • AI Agent Tutorials

  • AI Coding Tutorials

  • AI Model

  • AI Agent Papers

  • Chatbot

  • AI Dataset

  • Machine Learning Foundation Books

  • AI Train

  • AI Deploy

  • AI Client

  • Machine Learning Foundation Papers

  • Machine Learning Foundation Tutorials

  • AI Image Demos

  • AI Agent

  • Large Language Model Tutorials

  • Large Language Model Papers

  • Machine Learning Engineering Papers

  • Computer Vision Tutorials

  • Computer Vision Papers

  • Natural Language Processing Papers

  • Reinforcement Learning Papers

  • Speech Technology Papers

  • AI API

  • AI Coding

  • AI Image

  • AI Video

  • MLOps

  • MCP Client

  • MCP Server

  • AI Video Papers

  • AI Audio

  • AI Others

  • AI Infra

  • Embodied AI

Reinforcement Learning Papers·2026
Icon for item

Beyond Teacher Assignment: Domain-Normalized Multi-Teacher On-Policy Distillation

Xin Li, Hao Jiang +7

Normalizes each domain's teacher feedback spread during multi-teacher on-policy distillation so that no domain (e.g., instruction following) overwhelms others, improving student recovery of specialist skills and raising average scores across benchmarks.

#distillation#rl#routing#LLM#paper+2
Natural Language Processing Papers·2026
Icon for item

Improving Test-Time Scaling with Adaptive Looped Transformers

Yichen You, Tianyu Fu +5·Tsinghua University, Yale University

Improves test-time scaling of looped transformers by adaptively assigning extra recurrent iterations to tokens that benefit most. TaH2 is a post-training method that jointly trains an iteration decider with the backbone using lookahead depth supervision, boosting the accuracy–compute slope and peak accuracy on challenging benchmarks.

#LLM#NLP#paper#reasoning#transformers+6
Large Language Model Papers·2026
Icon for item

E-MoE: Enhanced Mixture-of-Experts for Non-Factorized Diffusion Language Models

Arseny Ivanov, Alexander Kolesov +3·Affiliation: AXXX, Moscow, Russia, Affiliation: Applied AI Institute, Moscow, Russia

Models sequence generation by unmasking multiple tokens per denoising step and replaces a factorized reverse process with a mixture over discrete routing-based latents from an MoE backbone; improves few-step sampling quality without increasing active parameters.

#moe#nlp#LLM#paper#routing+1
Large Language Model Papers·2026
Icon for item

Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It

Zehao Jin, Ruixuan Deng +1·Georgia Institute of Technology

Demonstrates that pretrained transformers typically use only ~1–3 lines of depth to follow reference chains, and that a task‑trained rank‑8 LoRA applied at one early layer (with all other weights frozen) can extend reference‑following to dozens or hundreds of lines while adding only a few ten‑thousand parameters.

#lora#qwen#transformers#LLM#NLP+4
Large Language Model Papers·2026
Icon for item

STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization

Bingchen Yao, Haobo Xu +7·City University of Hong Kong, Corresponding Author +5

Compresses recurrent states in linear-attention LLMs via spatial–temporal post-training quantization, allocating bits by error lifetime and per-row impact to preserve accuracy (6-bit ≈ FP32) while cutting serving memory up to 68.7%.

#quantization#qwen#llm#ai-inference#ai-serving+4
Large Language Model Papers·2026
Icon for item

Does Learning Protein Folding Generalize to Broader Reasoning?

Yong Liu, Zhanpeng Shi +5·Shanghai Jiao Tong University, Fudan University +2

Turns solved protein structures into FoldingCorpus and Fold2Reason — a post-training recipe that supervises an LLM with discrete structural Q&A plus continuous 3D geometry to improve spatial, graph and scientific reasoning; reports consistent gains across 10 benchmarks.

#paper#LLM#reasoning#qwen#biology+4
Reinforcement Learning Papers·2026
Icon for item

Make Sparse Rewards Count: Density-Aware Reward Aggregation for Multi-Reward RL

Tong Zheng, Skylar Zhai +8·University of Chinese Academy of Sciences, University of Minnesota Twin Cities +6

Calibrates multi-reward reinforcement learning by adaptively upweighting infrequently active rewards per rollout batch, so sparse objectives provide stronger signals when they matter. Proposes an inverse-square-root density correction and shows faster learning on tool-calling and math-reasoning tasks.

#RL#LLM#NLP#paper#github+1
Computer Vision Papers·2026
Icon for item

UniEvo-VL: An On-policy Self-Distillation Training Recipe for Multimodal Model Self-improvement

Fang Wu, Da Xing +17·Stanford University, Johns Hopkins University +8

Uses a multimodal model's own critiques as privileged context and applies on-policy self-distillation over diffusion sampling trajectories to internalize corrective guidance, improving text-to-image generation without an external teacher; shows measurable gains on GenEval and GenEval2.

#distillation#multimodal#qwen#evaluation#vision+2
AI Agent Papers·2026
Icon for item

False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents

Meijia Chen, Hao Li +13

Analyzes how proposer–solver loops in self-evolving search agents can develop shared errors (co-cheating) that inflate internal rewards; introduces Multi-Sample Verification and CrossFit (cross-fitted scoring with partitioned sources) to reduce false agreement and improve downstream search performance.

#ai-agent#web-search#qwen#RAG#retrieval+5
Reinforcement Learning Papers·2026
Icon for item

Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics

Maoqi Liu, Junwei He +6·Beijing University of Posts and Telecommunications, ByteDance

Demonstrates that additive weighted aggregation in rubric-based reinforcement learning enables reward hacking—models improve rubric coverage but produce worse answers—and proposes ProRubric, an offline regrouping that aggregates checklist criteria into conjunctive protocol-level dimensions to raise appropriateness without losing coverage.

#RL#LLM#evaluation#research#paper+2
Natural Language Processing Papers·2026
Icon for item

Hierarchical Continuous Diffusion Language Models

Hui Ren, Zihan Li +3·University of Illinois Urbana-Champaign, Amazon.com, Inc.

Couples a continuous latent diffusion trajectory with discrete token readouts so tokens are read from the latent at every step and fed back as a scaffold, reducing token-independence in parallel decoding and improving structured-task accuracy and LM perplexity.

#paper#NLP#LLM#diffusers#transformers+2
Large Language Model Papers·2026
Icon for item

ALoDLM: Adaptively Looped Diffusion Language Models

Liancheng Fang, Zhuowei Li +11

Introduces a token-adaptive latent recurrence for diffusion language models that allocates computation per token during denoising. Key features: iterative latent refinement, discrete-feedback commits, and learned token-wise schedules to improve parallel decoding quality-efficiency trade-offs.

#paper#LLM#NLP#diffusers#transformers+3
  • Previous
  • 1
  • More pages
  • 13
  • 14
  • 15
  • More pages
  • 18
  • Next