思想与社会2025 · 分享于 Hacker News
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL
探索《DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL》的原文与 Hacker News 社区讨论，查看来源并加入阅读清单。
从论文，走向讨论
看看社区如何讨论这个想法。
来自 Hacker News 的原始讨论。不同的视角、疑问，以及值得继续思考的线索。
最多展示 100 条相关讨论。论文可能出现在评论中，并非原帖投稿内容。
在讨论中发现的引用
按 HN 条目顺序展示最近的 50 条引用来源。
评论 #48879100 评论 #48422249 评论 #47900411 评论 #47653535 评论 #46701709 评论 #46578762 评论 #46530363 评论 #46096548 评论 #45121239 评论 #44265349 评论 #44265256 评论 #44246502 评论 #43713920 评论 #42954882 评论 #42904497 评论 #42887752 评论 #42877272 评论 #42873399 评论 #42873275 评论 #42856385 评论 #42849687 评论 #42845694 评论 #42845671 评论 #42845019 评论 #42843106 评论 #42833467 评论 #42829571 评论 #42813026 评论 #42804083 评论 #42804078
Ygradus_ad ·
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL
1,351 赞同1,056 评论
Yreissbaker ·
DeepSeek releases Janus Pro, a text-to-image generator [pdf]
912 赞同674 评论
Ytkellogg ·
S1: A $6 R1 competitor?
851 赞同416 评论
Ytalhof8 ·
Exposed DeepSeek database leaking sensitive information, including chat history
707 赞同474 评论
Yeigenvalue ·
The impact of competition and DeepSeek on Nvidia
655 赞同479 评论
Yamrrs ·
The Illustrated DeepSeek-R1
578 赞同118 评论
Yjxmorris12 ·
RLHF Book
479 赞同37 评论
YJnBrymn ·
The Bitter Lesson Is Misunderstood
369 赞同251 评论
Yfirloop ·
The Gentle Singularity
268 赞同527 评论
Ypella ·
Emerging reasoning with reinforcement learning
248 赞同211 评论
Ybuyucu ·
Llama.cpp supports Vulkan. why doesn't Ollama?
217 赞同228 评论
YOlshansky ·
Citations on the Anthropic API
177 赞同61 评论
Ysemking ·
Berkeley Researchers Replicate DeepSeek R1's Core Tech for Just $30: A Small Mod
163 赞同66 评论
Ycubefox ·
The inefficiency of RL, and implications for RLVR progress
120 赞同48 评论
Yvincent_s ·
DeepSeek and the Effects of GPU Export Controls
50 赞同51 评论
Yneamar ·
In Two Moves, AlphaGo and Lee Sedol Redefined the Future (2016)
34 赞同23 评论
Yatomic128 ·
AI industry insiders launch site to poison the data that feeds them
32 赞同4 评论
Yamrrs ·
How Deepseek R1 Was Trained
31 赞同3 评论
YjohnnyApplePRNG ·
Ask HN: Can anybody clarify why OpenAI reasoning now shows non-English thoughts?
22 赞同41 评论
YBrajeshwar ·
Cognitive Reasoning Agents and the Extended Information Filter
14 赞同4 评论
Ysabas_ge ·
DeepSeek temporarily limits registration to Chinese mobile numbers
13 赞同11 评论
Yjamie-simon ·
There Will Be a Scientific Theory of Deep Learning
8 赞同0 评论
Ynsoonhui ·
Alibaba to ban Claude Code in workplace over alleged backdoor risks, source says
8 赞同2 评论
Yksymph ·
DeepSeek-R1 paper updated from 22 pages to 86 with additional details
4 赞同1 评论
Y0xkato ·
How LLMs Work
4 赞同0 评论
YScramblejams ·
Nanolang: A tiny experimental language designed to be targeted by coding LLMs
2 赞同0 评论
Ydesideratum ·
Nanocode: The best Claude Code that $200 can buy in pure JAX on TPUs
2 赞同0 评论