Hacker News with Generative AI: Reinforcement Learning

The State of Reinforcement Learning for LLM Reasoning (sebastianraschka.com)
A lot has happened this month, especially with the releases of new flagship models like GPT-4.5 and Llama 4. But you might have noticed that reactions to these releases were relatively muted. Why? One reason could be that GPT-4.5 and Llama 4 remain conventional models, which means they were trained without explicit reinforcement learning for reasoning.
Generative AI, Reinforcement Learning
7 points by yaiml 1 day ago | 0 comments
Does RL Incentivize Reasoning in LLMs Beyond the Base Model? (limit-of-rlvr.github.io)
Recent breakthroughs in reasoning-focused large language models (LLMs) like OpenAI-o1, DeepSeek-R1, and Kimi-1.5 have largely relied on Reinforcement Learning with Verifiable Rewards (RLVR), which replaces human annotations with automated rewards (e.g., verified math solutions or passing code tests) to scale self-improvement. While RLVR enhances reasoning behaviors such as self-reflection and iterative refinement, we challenge a core assumption:
Artificial Intelligence, Reinforcement Learning
84 points by leodriesch 4 days ago | 38 comments
The State of Reinforcement Learning for LLM Reasoning (sebastianraschka.com)
A lot has happened this month, especially with the releases of new flagship models like GPT-4.5 and Llama 4. But you might have noticed that reactions to these releases were relatively muted. Why? One reason could be that GPT-4.5 and Llama 4 remain conventional models, which means they were trained without explicit reinforcement learning for reasoning.
Reinforcement Learning, Generative AI, AI Research
9 points by jonbaer 5 days ago | 0 comments
Welcome to the Era of Experience [pdf] (googleapis.com)
Artificial Intelligence, Reinforcement Learning, Machine Learning
114 points by Siah 6 days ago | 52 comments
Skywork-OR1: new SOTA 32B thinking model with open weight (github.com/SkyworkAI)
✊ Unleashing the Power of Reinforcement Learning for Math and Code Reasoners 🤖
Artificial Intelligence, Machine Learning, Open Source, Reinforcement Learning
145 points by naomiclarkson 13 days ago | 28 comments
DeepCoder: An Open-Source 14B Coder at O3-Mini Level (together.ai)
Through a joint collaboration between the Agentica team and Together AI, we release DeepCoder-14B-Preview, a code reasoning model finetuned from Deepseek-R1-Distilled-Qwen-14B via distributed RL. It achieves an impressive 60.6% Pass@1 accuracy on LiveCodeBench (+8% improvement), matching the performance of o3-mini-2025-01-031 (Low) and o1-2024-12-17 with just 14B parameters. We’ve open-sourced our dataset, code, training logs, and systems optimizations for everyone to progress on scaling and accelerating intelligence with RL.
Open Source, Code Generation, Reinforcement Learning, Machine Learning
31 points by tosh 17 days ago | 4 comments
Can reinforcement learning for LLMs scale beyond math and coding tasks? Probably (arxiv.org)
Reinforcement learning with verifiable rewards (RLVR) has demonstrated significant success in enhancing mathematical reasoning and coding performance of large language models (LLMs), especially when structured reference answers are accessible for verification.
Reinforcement Learning, Artificial Intelligence, Computer Science
6 points by GabrielBianconi 18 days ago | 4 comments
DeepSeek: Inference-Time Scaling for Generalist Reward Modeling (arxiv.org)
Reinforcement learning (RL) has been widely adopted in post-training for large language models (LLMs) at scale.
Reinforcement Learning, Artificial Intelligence, Machine Learning
163 points by tim_sw 22 days ago | 35 comments
Search-R1: Training LLMs to Reason and Leverage Search Engines with RL (arxiv.org)
Efficiently acquiring external knowledge and up-to-date information is essential for effective reasoning and text generation in large language models (LLMs).
AI, Reasoning, Search Engines, Reinforcement Learning
101 points by jonbaer 23 days ago | 13 comments
Scaling Up Reinforcement Learning for Traffic Smoothing (bair.berkeley.edu)
We deployed 100 reinforcement learning (RL)-controlled cars into rush-hour highway traffic to smooth congestion and reduce fuel consumption for everyone.
Reinforcement Learning, Traffic Management, AI, Transportation, Research
71 points by saeedesmaili 24 days ago | 21 comments
Launch HN: Augento (YC W25) – Fine-tune your agents with reinforcement learning (ycombinator.com)
Hi HN, we’re the cofounders of Augento (https://augento.ai/). We’re building Deepseek R1-like fine-tuning as a service. You connect your agent, tell us when it’s right or wrong, and we deliver an LLM optimized for that agent.
Machine Learning, Reinforcement Learning, Startups, AI, SaaS
101 points by lmeierhoefer 26 days ago | 60 comments
I Built Faster Reinforcement Learning in C# Solo Than Teams Did with Python (rlmatrix.com)
The question comes relentlessly: “Why build reinforcement learning in C#?” Behind this query lies an unspoken assumption that serious machine learning happens exclusively in Python. This perspective reveals a fundamental disconnect between academic ML researchers with their sprawling Python scripts and those of us solving real industrial problems.
Reinforcement Learning, Programming Languages, C#, Python, Machine Learning
13 points by adriansieradzki 28 days ago | 2 comments
A (Long) Peek into Reinforcement Learning (lilianweng.github.io)
A couple of exciting news in Artificial Intelligence (AI) has just happened in recent years. AlphaGo defeated the best professional human player in the game of Go. Very soon the extended algorithm AlphaGo Zero beat AlphaGo by 100-0 without supervised learning on human knowledge. Top professional game players lost to the bot developed by OpenAI on DOTA2 1v1 competition. After knowing these, it is pretty hard not to be curious about the magic behind these algorithms — Reinforcement Learning (RL).
Artificial Intelligence, Machine Learning, Reinforcement Learning
160 points by Brysonbw 34 days ago | 16 comments
Understanding R1-Zero-Like Training: A Critical Perspective (github.com/sail-sg)
To understand R1-Zero-like training, we critically examine two core components: base models and reinforcement learning. We highlight our findings below.
Machine Learning, Reinforcement Learning, Research, Artificial Intelligence
160 points by pama 35 days ago | 21 comments
Hunyuan T1 Mamba Reasoning model beats R1 on speed and metrics (tencent.github.io)
Reinforcement learning has pioneered a new Scaling paradigm in the post-training phase of large language models, a breakthrough that is increasingly attracting attention from the industry.
Artificial Intelligence, Reinforcement Learning
26 points by vessenes 35 days ago | 2 comments
Legged Locomotion Meets Skateboarding (umich-curly.github.io)
This paper introduces Discrete-time Hybrid Automata Learning (DHAL), a framework using on-policy Reinforcement Learning to identify and execute mode-switching without trajectory segmentation or event function learning.
Robotics, Reinforcement Learning, Machine Learning
156 points by jam 36 days ago | 37 comments
Mathematical Foundations of Reinforcement Learning (github.com/MathFoundationRL)
This textbook has received 5,000+ stars! Glad that it is helpful to many readers.
Reinforcement Learning, Mathematics, Computer Science, Education
424 points by ibobev 47 days ago | 43 comments
Reinforcement Learning in less than 400 lines of C (github.com/antirez)
This code implements a neural network that learns to play tic-tac-toe using reinforcement learning, just playing against a random adversary, in under 400 lines of C code, without any external library used.
Reinforcement Learning, Programming, C Programming, Tic-Tac-Toe
10 points by antirez 47 days ago | 4 comments
Show HN: Llama-8B Teaches Itself Baby Steps to Deep Research Using RL (github.com/dCaples)
Autonomously train research-agent LLMs on custom data using reinforcement learning and self-verification.
Reinforcement Learning, Research, Artificial Intelligence, Open Source
39 points by diegocaples 47 days ago | 3 comments
All Roads Lead to Likelihood: The Value of Reinforcement Learning in Fine-Tuning (arxiv.org)
From a first-principles perspective, it may seem odd that the strongest results in foundation model fine-tuning (FT) are achieved via a relatively complex, two-stage training procedure.
Machine Learning, Reinforcement Learning, Fine-Tuning, Foundation Models
3 points by gkswamy98 49 days ago | 0 comments
Using GRPO to Beat o1, o3-mini and R1 at “Temporal Clue” (openpipe.ai)
In this post we’ll discuss how we used Group Relative Policy Optimization (GRPO) to surpass R1, o1, o3-mini, and come within a couple percentage points of Sonnet 3.7 on a reasoning-heavy game called “temporal clue”, while being over 100x cheaper to run at inference time. We’ll include specific lessons learned about task design and hyperparameters we’ve found to work well. And finally, we share the training recipe we used to achieve these results, built on top of torchtune.
Artificial Intelligence, Reinforcement Learning, Game Development, Optimization
199 points by kcorbitt 51 days ago | 55 comments
QwQ-32B: Embracing the Power of Reinforcement Learning (qwenlm.github.io)
Scaling Reinforcement Learning (RL) has the potential to enhance model performance beyond conventional pretraining and post-training methods.
Reinforcement Learning, Machine Learning, Generative AI
480 points by nwjsmith 52 days ago | 169 comments
RoboPianist: Dexterous Piano Playing with Deep Reinforcement Learning (2023) (kzakka.com)
We train anthropomorphic robot hands to play the piano using deep RL and release a simulated benchmark and dataset to advance high-dimensional control.
Robotics, Reinforcement Learning, Music, Artificial Intelligence, Machine Learning
142 points by bemmu 58 days ago | 53 comments
Competitive Programming with Large Reasoning Models (arxiv.org)
We show that reinforcement learning applied to large language models (LLMs) significantly boosts performance on complex coding and reasoning tasks.
Competitive Programming, Reinforcement Learning, Artificial Intelligence
16 points by t55 73 days ago | 1 comments
DeepScaleR: Surpassing O1-Preview with a 1.5B Model by Scaling RL (notion.site)
DeepScaleR is an open-source project to fully democratize reinforcement learning (RL) for LLMs and reproduce DeepSeek R1 and OpenAI O1/O3 at scale on real tasks.
Reinforcement Learning, Open Source
322 points by sijuntan 74 days ago | 127 comments
DeepScaleR: Surpassing O1-Preview with a 1.5B Model by Scaling RL (notion.site)
Deep Learning, Reinforcement Learning
19 points by mluo 75 days ago | 0 comments
Craftax: (Crafter and NetHack) RL Environment in Jax (github.com/MichaelTMatthews)
Craftax is an RL environment written entirely in JAX. Craftax reimplements and significantly extends the game mechanics of Crafter, taking inspiration from roguelike games such as NetHack.
Reinforcement Learning, Game Development, Python, Jax
14 points by lnyan 75 days ago | 2 comments
The Differences Between Direct Alignment Algorithms Are a Blur (arxiv.org)
Direct Alignment Algorithms (DAAs) simplify language model alignment by replacing reinforcement learning (RL) and reward modeling (RM) in Reinforcement Learning from Human Feedback (RLHF) with direct policy optimization.
Reinforcement Learning, Artificial Intelligence, Machine Learning
8 points by t55 76 days ago | 0 comments
There may not be aha moment in R1-Zero-like training (notion.site)
Machine Learning, Artificial Intelligence, Reinforcement Learning
78 points by qianli_cs 78 days ago | 8 comments
R1 Computer Use (github.com/agentsea)
r1-computer-use is an experimental project that applies large-scale Reinforcement Learning techniques similar to DeepSeek-R1 to computer usage scenarios.
Reinforcement Learning, Machine Learning, Experimental Projects, Computer Science
104 points by mountainriver 79 days ago | 57 comments