Name: [PDG 464] Kimi K1.5 Scaling RL with LLMs
Start: 2025-11-25T20:00:00+01:00
End: 2025-11-25T22:00:00+01:00

**Link to article**: https://arxiv.org/pdf/2501.12599
**Title:** KIMI K1.5: SCALING REINFORCEMENT LEARNING WITH LLMS
**Track:** RL
**Content**: The authors present Kimi k1.5, a multi-modal LLM trained with a simplified but large-scale reinforcement learning framework (no MCTS, value functions, or process rewards) that leverages long-context scaling and improved policy optimization to effectively generate its own training data through exploration. This approach achieves state-of-the-art reasoning performance comparable to OpenAI’s o1 on long-CoT benchmarks and, via “long2short” methods, yields short-CoT models that substantially outperform existing systems like GPT-4o and Claude Sonnet 3.5 on math and coding tasks.
**Slack link**: ml-ka.slack.com, channel: #pdg. Please join us -- if you cannot join, please message us here or to mlpaperdiscussiongroupka@gmail.com.

In the Paper Discussion Group (PDG) we discuss recent and fundamental papers in the area of machine learning on a weekly basis. If you are interested, **please** **read the paper beforehand** and join us for the discussion. If you have not fully understood the paper, you can still participate – everyone is welcome! You can join the discussion or simply listen in. The discussion is in German or English depending on the participants.

DavidFarago

AI Paper Discussion Group Karlsruhe

Technology

Data Science

Neural Networks

Deep Learning

Natural Language Processing

Artificial Intelligence

Machine Learning

Academics

Intellectual Discussions

Cristi Alexa

Mario Zachmann

Lionel Chiron

Leonard Plotkin

r-stl

Christoph Michel

Vivi

[PDG 464] Kimi K1.5 Scaling RL with LLMs

Online event

Teilen

AI Paper Discussion Group Karlsruhe

[PDG 464] Kimi K1.5 Scaling RL with LLMs

AI Paper Discussion Group Karlsruhe

Details

Mitglieder interessieren sich auch für