Papers119

#GRPO

Multi-hop Reasoning via Early Knowledge Alignment

Yuxin Wang, Shicheng Fang et al.Dec 23arXiv

This paper adds a tiny but powerful step called Early Knowledge Alignment (EKA) to multi-step retrieval systems so the model takes a quick, smart look at relevant information before it starts planning.

#Retrieval-Augmented Generation#Iterative RAG#Multi-hop Reasoning

Memory-T1: Reinforcement Learning for Temporal Reasoning in Multi-session Agents

Intermediate

Yiming Du, Baojun Wang et al.Dec 23arXiv

Memory-T1 teaches chatty AI agents to keep track of when things happened across many conversations.

#temporal reasoning#multi-session dialogue#reinforcement learning

GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators

Intermediate

Jiacheng Guo, Ling Yang et al.Dec 22arXiv

GenEnv is a training system where a student AI and a teacher simulator grow together by exchanging tasks and feedback.

#GenEnv#co-evolutionary learning#difficulty-aligned curriculum

VA-$π$: Variational Policy Alignment for Pixel-Aware Autoregressive Generation

Intermediate

Xinyao Liao, Qiyuan He et al.Dec 22arXiv

Autoregressive (AR) image models make pictures by choosing tokens one-by-one, but they were judged only on picking likely tokens, not on how good the final picture looks in pixels.

#autoregressive image generation#tokenizer–generator alignment#pixel-space reconstruction

Reasoning Palette: Modulating Reasoning via Latent Contextualization for Controllable Exploration for (V)LMs

Intermediate

Rujiao Long, Yang Li et al.Dec 19arXiv

Reasoning Palette gives a language or vision-language model a tiny hidden “mood” (a latent code) before it starts answering, so it chooses a smarter plan rather than just rolling dice on each next word.

#Reasoning Palette#latent contextualization#VAE

Reinforcement Learning for Self-Improving Agent with Skill Library

Intermediate

Jiongxiao Wang, Qiaojing Yan et al.Dec 18arXiv

This paper teaches AI agents to learn new reusable skills and get better over time by using reinforcement learning, not just prompts.

#Reinforcement Learning#Skill Library#Sequential Rollout

Turn-PPO: Turn-Level Advantage Estimation with PPO for Improved Multi-Turn RL in Agentic LLMs

Intermediate

Junbo Li, Peng Zhou et al.Dec 18arXiv

Turn-PPO is a new way to train chatty AI agents that act over many steps, by judging each conversation turn as one whole action instead of judging every single token.

#Turn-PPO#multi-turn reinforcement learning#agentic LLMs

Differences That Matter: Auditing Models for Capability Gap Discovery and Rectification

Intermediate

Qihao Liu, Chengzhi Mao et al.Dec 18arXiv

AuditDM is a friendly 'auditor' model that hunts for where vision-language models get things wrong and then creates the right practice to fix them.

#AuditDM#model auditing#cross-model divergence

RePlan: Reasoning-guided Region Planning for Complex Instruction-based Image Editing

Intermediate

Tianyuan Qu, Lei Ke et al.Dec 18arXiv

RePlan is a plan-then-execute system that first figures out exactly where to edit in a picture and then makes clean changes there.

#instruction-based image editing#vision–language model (VLM)#diffusion model

JustRL: Scaling a 1.5B LLM with a Simple RL Recipe

Intermediate

Bingxiang He, Zekai Qu et al.Dec 18arXiv

JustRL shows that a tiny, steady recipe for reinforcement learning (RL) can make a 1.5B-parameter language model much better at math without fancy tricks.

#Reinforcement Learning#GRPO#Policy Entropy

Skyra: AI-Generated Video Detection via Grounded Artifact Reasoning

Intermediate

Yifei Li, Wenzhao Zheng et al.Dec 17arXiv

Skyra is a detective-style AI that spots tiny visual mistakes (artifacts) in videos to tell if they are real or AI-generated, and it explains its decision with times and places in the video.

#AI-generated video detection#artifact reasoning#multimodal large language model

Can LLMs Guide Their Own Exploration? Gradient-Guided Reinforcement Learning for LLM Reasoning

Intermediate

Zhenwen Liang, Sidi Lu et al.Dec 17arXiv

This paper teaches large language models (LLMs) to explore smarter by listening to their own gradients—the directions they would update—rather than chasing random variety.

#gradient-guided reinforcement learning#GRL#GRPO

6 7 8 9 10