[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-ettin-reranker-6-sizes-17m-1b-apache2":3,"topics-all":36,"news-related-59182340-526a-427b-b290-184534d703a3":55},{"id":4,"title":5,"summary":6,"content":6,"original_url":7,"source_id":8,"tags":9,"translations":23,"news_slug":29,"published_at":30,"created_at":31,"modified_at":32,"is_published":33,"publish_type":34,"image_url":13,"view_count":35},"59182340-526a-427b-b290-184534d703a3","Ettin Reranker 系列登场：六款不同尺寸的开源 SOTA 跨编码器，小到 17M 也能越级打","Hugging Face 工程师 Tom Aarsen 5 月 19 日发布 Ettin Reranker 系列，六个 Apache 2.0 开源 Sentence Transformers CrossEncoder，从 17M 到 1B 全档覆盖。所有模型在 Ettin ModernBERT 编码器上以 pointwise MSE 蒸馏自 1.54B 的 mxbai-rerank-large-v2，训练数据 ~1.43 亿 (query, document, score) 三元组全部公开。配合 Flash Attention 2 + bf16 在 H100 上做推理，速度比默认加载快 1.7x–8.3x。基准方面，1B 模型在 MTEB(eng, v2) Retrieval 上以 0.6114 与 1.54B 教师持平 (差 0.0001)，NanoBEIR 上差 0.008。150M 规模在 MTEB 上以 0.5994 反超 Qwen3-Reranker-0.6B (596M)。最小的 17M 即可在 MTEB 上以 0.5576 击败 33M 的 ms-marco-MiniLM-L12-v2 (0.5066)，32M 在 MTEB 上以 0.5779 击败 568M 的 bge-reranker-v2-m3 (0.5526)，17x 参数差下实现反超。架构层面采用 unpadded 注意力、RoPE、GeGLU 与 4 模块分类头，CLS 池化优于 mean 池化，得益于 ModernBERT 每三层一次的全局注意力。所有模型支持 8192 token 上下文，可直接 drop-in 替换现有 retrieve-then-rerank 栈中的 MiniLM 系列重排序器。","https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fettin-reranker","24d5c6c5-6573-4180-a1fd-f1459842d1af",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"120fa59a-ff6f-4537-9bf5-f818df636a0e","benchmark",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"0ef8513a-0a26-42f0-b6f9-5b6dadded45c","efficiency",{"id":18,"name":19,"slug":19,"description":13,"color":13},"7e89b5cc-57db-4f37-bc6d-28919a73931c","model-release",{"id":21,"name":22,"slug":22,"description":13,"color":13},"b9bd9039-fcdb-41a8-b85b-fc1587def2b9","open-source",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"53a47e8b-ce91-4eab-87ec-3dd0f8500c4e","en","Ettin rerankers: six open SOTA cross-encoders, 17M punches up","Hugging Face engineer Tom Aarsen released the Ettin Reranker series on May 19 — six Apache 2.0 open-source Sentence Transformers CrossEncoders, covering the full range from 17M to 1B. All models are distilled pointwise MSE from the 1.54B mxbai-rerank-large-v2 on the Ettin ModernBERT encoder, with all ~143M (query, document, score) triples in the training data publicly released. Combined with Flash Attention 2 + bf16 inference on H100, the speed is 1.7×-8.3× faster than default loading. On the benchmark side, the 1B model ties the 1.54B teacher on MTEB (eng, v2) Retrieval at 0.6114 (0.0001 difference), and lags by 0.008 on NanoBEIR. The 150M size beats Qwen3-Reranker-0.6B (596M) on MTEB with 0.5994. The smallest 17M beats ms-marco-MiniLM-L12-v2 (33M, 0.5066) on MTEB with 0.5576, the 32M beats bge-reranker-v2-m3 (568M, 0.5526) on MTEB with 0.5779, achieving the upset with a 17× parameter gap. Architecturally it uses unpadded attention, RoPE, GeGLU, and a 4-module classification head, with CLS pooling beating mean pooling, benefiting from ModernBERT's once-every-three-layers global attention. All models support 8192 token context, and can directly drop-in replace the MiniLM series rerankers in existing retrieve-then-rerank stacks.","ettin-reranker-6-sizes-17m-1b-apache2","2026-06-10T06:30:00Z","2026-06-10T06:27:50.233370Z","2026-08-19T02:08:40.142862Z",true,"agent",178,[37,46],{"slug":38,"tag_slug":38,"title_zh":39,"title_en":40,"intro_zh":41,"intro_en":42,"id":43,"is_active":33,"created_at":44,"modified_at":45},"ai-for-science","AI for Science 2026：从 UniPert 到 GPT-Rosalind 的硬核进化","AI for Science 2026: from UniPert to GPT-Rosalind","生命科学、化学材料、物理世界模型——AI 正在从\"语言工具\"变成\"实验伙伴\"。本专题收录 AI 在三大科学方向的关键节点：UniPert 统一基因与化学扰动空间、GPT-Rosalind 端到端生命科学推理、达摩院 AI 智能体 28 小时找到 4 种超导新材料、Anthropic Claude Science 把工作台做成标准品。","From language tool to lab partner — AI is reshaping life sciences, chemistry\u002Fmaterials, and physical world models. This topic covers the key milestones: UniPert unifying genetic-chemical perturbation spaces, GPT-Rosalind's end-to-end life-sciences reasoning, DAMO's AI agent discovering 4 superconducting materials in 28 hours, and Anthropic's Claude Science workbench going mainstream.","988a4300-5fab-41c4-b5d8-63711a2dc757","2026-09-10T01:34:15.296649Z","2026-09-10T01:34:15.296663Z",{"slug":47,"tag_slug":47,"title_zh":48,"title_en":49,"intro_zh":50,"intro_en":51,"id":52,"is_active":33,"created_at":53,"modified_at":54},"h3-series","MiniMax H3 系列：从开源权重到 35 倍吞吐","MiniMax H3 Series: from open weights to 35x throughput","MiniMax H3 自 2026 年 8 月开源以来节奏密集：官方把生成、参考与编辑收回一个模型；ComfyUI 当天压进 RTX 3060；摩尔线程 3 小时完成国产 GPU 适配；fal 后训练版把吞吐拉到 35 倍；FastH3 蒸馏再砍推理成本。本专题持续追踪 H3 的发布—开源—蒸馏—部署全链路。","Since MiniMax open-sourced H3 in August 2026 the pace has been relentless: one unified omni-modal model, same-day ComfyUI support down to an RTX 3060, a 3-hour Day-0 port to Moore Threads GPUs, fal's post-trained H3 Max at 35x throughput, and FastH3 distillation cutting inference cost further. This topic tracks the full H3 chain — release, open weights, distillation, deployment.","83ef0daa-3c31-4cb3-86ed-e5ee58654d5f","2026-09-08T07:33:19.942193Z","2026-09-08T07:33:19.942209Z",{"items":56},[57,62,67,72,77,82],{"id":58,"title":59,"news_slug":60,"published_at":61},"2e27016d-b90e-45c7-825a-41fd1e435c80","JHU 新研究:组合持续学习机制,百任务记忆留存从 1.2% 提到 34.9%","compose-cl-long-horizon-memorization","2026-09-16T15:10:00+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"4c4a2a9e-f69b-4985-bd42-97ab2ef4e2ac","Spark-X2.5-4B 开源:4B 跑 1M 上下文,22 项基准打 9B 级 Qwen3.5","spark-x2-5-4b-apache-open-source","2026-09-16T01:30:00+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"30fca629-bace-4832-9789-b44aa8c8989d","学生团队从零训出开源 7B 模型 ZGCM-1:数学推理硬刚 235B 前沿","zgcm-1-open-7b-foundation-model","2026-09-15T19:10:00+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"2731ed1c-17c3-4d85-9174-983cf50743e3","地铁售票机上的 AI 大考:2.6GB 端侧模型 91.32 分超 GPT-5.6,规则基线也拿 84.6","metrollm-bench-transit-kiosk-llm","2026-09-12T23:08:18+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"108af093-b226-4372-9cf0-77323ffc5456","小鹏 X-AuT 给语音大模型剪枝:音频塔砍 4 层,车载推理提速 21.4%","xpeng-x-aut-audio-encoder-pruning","2026-09-12T19:06:47+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"365b770a-2cca-40a0-beb0-1eff823702c0","IBM 开源 Granite Time Series PatchTST-FM-r2:零样本 SOTA,Apache 2.0 商用许可","ibm-granite-patchtst-fm-r2-zero-shot-apache","2026-09-12T11:00:00+00:00"]