Experimental browser for the Atmosphere
Delighted to see BigGen Bench paper receive the πbest paper award πat NAACL 2025! BigGen Bench introduces fine-grained, scalable, & human-aligned evaluations: π 77 hard, diverse tasks π οΈ 765 exs w/ ex-specific rubrics π More human-aligned than previous rubrics π 10 languages, by native speakers 1/
May 6, 2025, 1:49 PM
{ "uri": "at://did:plc:evvussoazdkvsld475dfbuci/app.bsky.feed.post/3loizm7whl22b", "cid": "bafyreicqejkjea7bwswh36wm6nu7lqpsca3qmk33mvpvryiamurrk54n6i", "value": { "text": "Delighted to see BigGen Bench paper receive the πbest paper award πat NAACL 2025!\n\nBigGen Bench introduces fine-grained, scalable, & human-aligned evaluations:\n\nπ 77 hard, diverse tasks\nπ οΈ 765 exs w/ ex-specific rubrics\nπ More human-aligned than previous rubrics\nπ 10 languages, by native speakers\n\n1/", "$type": "app.bsky.feed.post", "embed": { "$type": "app.bsky.embed.images", "images": [ { "alt": "", "image": { "$type": "blob", "ref": { "$link": "bafkreihdaormwb6gryciy73s4l26lyfz6ojlbnmcjgpwjqvbo4b7rvrqwa" }, "mimeType": "image/jpeg", "size": 391521 }, "aspectRatio": { "width": 2000, "height": 1128 } } ] }, "langs": [ "en" ], "createdAt": "2025-05-06T13:49:57.366Z" } }