AI 与机器学习2025arxiv.org
DeepSeek:面向通用奖励建模的推理时扩展
DeepSeek: Inference-Time Scaling for Generalist Reward Modeling
进阶 · 约 50 分钟读完前置知识:RLHF 与奖励模型基础、强化学习基础、LLM 采样与推理
导读
这篇 DeepSeek 论文研究通用奖励模型(Reward Modeling)的推理时扩展问题,提出通过在推理阶段投入更多计算来提升奖励模型对 LLM 输出的评判质量,并介绍了相关方法(如 GRM)在推理时扩展下取得 SOTA 表现。
HN 关注点:DeepSeek 持续以开源方式发布 SOTA 级别研究,讨论焦点集中在开源实践与模型实际使用体验。
HN 评论精选
这几条评论里:支持 1 质疑 1 补充 1 亲历 1
支持赞赏 DeepSeek 的开源精神,并求推荐类似的公司
不仅惊叹于每篇论文都是 SOTA,更难得的是它以纯粹的开源定义开放成果,许可也很宽松。别把公司和国家的叙事过度绑定,流行媒体在强化仇恨以换取点击和金钱。政客之间或许互相敌视,但科学家乐于与志同道合者合作,竞争只在于可衡量的成就。讽刺的是,这家专注金融的公司愿意分享进展,而一些非营利和纯 AI 公司却把所有发现锁起来。有谁知道还有哪些像 DeepSeek 这样常发好论文的公司?我已关注 Mistral,希望丰富信息来源。
ALLTaken查看英文原文
Not jus being impressed that every paper coming out is SOTA, but also leads the way in being Open-Source in the pure definition of OSS, even with permissible licensing. Let's not confuse the company with the country by over-fitting a narrative. Popular media is reenforcing hatred or anything that sponsors them, especially to weaker groups. Less repercussions and more clicks/money to be made I guess. While Politicians may hate each other, Scientists love to work with other aspiring Scientists who have similar ambitions and the only competition is in achieving measurable success and the reward it means to the greater public. Without any bias, but it's genuinely admirable when companies release their sources to enable faster scientific progress cycles. It's ironic that this company is dedicated to finance, yet shares their progress, while non-profits and companies dedicated purely to AI are locking all knowledge about their findings from access. Are there other companies like DeepSeek that you know of that commonly release great papers? I am following Mistral already, but I'd love to enrich my sources of publications that I consume. Highly appreciated!
亲历亲历称 DeepSeek R1 文笔最佳
DeepSeek R1 是目前写散文最好的模型,超过 Grok-3、GPT-4o、o1-pro、o3、Claude 等。贴一段书里的文字让它按该风格续写,大多数模型的表现差得惊人。Grok-3 紧随其后,可能因为它背后其实就是改了几处的 DeepSeek R1。
resters查看英文原文
DeepSeek R1 is by far the best at writing prose of any model, including Grok-3, GPT-4o, o1-pro, o3, claude, etc. Paste in a snippet from a book and ask the model to continue the story in the style of the snippet. It's surprising how bad most of the models are. Grok-3 comes in a close second, likely because it is actually DeepSeek R1 with a few mods behind the scenes.
补充称赞论文术语使用更规范
很高兴看到 DeepSeek 使用了更正确、更地道的术语 "inference-time scaling",而不是 OpenAI 发明的那个别扭的 "test-time compute"。
mentalgear查看英文原文
Happy to see deekseek using the correct (and much more idiomatic) term "inference-time scaling", instead of the grotesque construction of "test-time compute" that openAI came up with.
质疑自述已从 DeepSeek 转回 Claude
有人知道我为什么对 DeepSeek 失去兴趣了吗?它刚发布时我和 Grok3 用了很多,但现在所有事情都退回用 Claude 了。
bilsbie查看英文原文
Any idea why I lost interest in deep seek? I used it and grok3 a whole bunch when they first came out but now I’ve fallen back to Claude for everything.
历次讨论
- 2025Hacker News 讨论帖163 分 · 35 评
同领域还值得读
DeepSeek-R1:通过强化学习激励 LLM 的推理能力
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL
研究如何用强化学习激励大语言模型的推理能力,提出 DeepSeek-R1,并探讨 RL 训练对模型推理表现的影响。
AI 与机器学习2025 · 1,351 分 · 1,056 评 · 引用 30进阶 · 约 80 分钟编辑精选1-bit LLM 时代:面向低成本计算的三值参数
The Era of 1-bit LLMs: ternary parameters for cost-effective computing
提出用三值参数 {-1, 0, 1} 替代 LLM 中的浮点权重,即所谓“1-bit LLM”。
AI 与机器学习2024 · 1,040 分 · 447 评 · 引用 19进阶 · 约 45 分钟编辑精选DeepSeek-V3.2:推进开放大语言模型的前沿
DeepSeek-v3.2: Pushing the frontier of open large language models
这是 DeepSeek-V3.2 的技术报告,延续了 DeepSeek 开源权重模型的路线,目标是让开放模型继续逼近甚至追平前沿闭源模型。
AI 与机器学习2025 · 982 分 · 465 评进阶 · 约 90 分钟编辑精选通过自我对弈与通用强化学习掌握国际象棋和将棋(AlphaZero)
Mastering Chess and Shogi by Self-Play with General Reinforcement Learning
DeepMind 的 AlphaZero 论文:用通用强化学习算法(自我博弈 + MCTS + 神经网络),从零开始仅数小时训练便在国际象棋和将棋上击败了世界顶级引擎 Stock…
AI 与机器学习2017 · 539 分 · 270 评 · 引用 13进阶 · 约 90 分钟编辑精选用强化学习发现更快的矩阵乘法算法
Discovering faster matrix multiplication algorithms with reinforcement learning
论文提出 AlphaTensor,用强化学习搜索矩阵乘法算法。
AI 与机器学习2022 · 518 分 · 111 评专业 · 约 90 分钟编辑精选