File size: 20,448 Bytes
00668ab |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 |
{
"search": [
"ACL 2025 Best paper",
"A Theory of Response Sampling in LLMs: Part Descriptive and Part Prescriptive",
"Fairness through Difference Awareness: Measuring Desired Group Discrimination in LLMs",
"Language Models Resist Alignment: Evidence From Data Compression",
"Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"
],
"result": {
"ACL 2025 Best paper": [
{
"page": 1,
"results": [
{
"snippet": "Best Paper · A Theory of Response Sampling in LLMs: Part Descriptive and Part Prescriptive · Fairness through Difference Awareness: Measuring Desired Group ...Read more",
"title": "Awards",
"url": "https://2025.aclweb.org/program/awards/"
},
{
"snippet": "Just now, the NSA paper by Liang Wenfeng from DeepSeek and the team led by Yang Yaodong from Peking University won the Best Paper Award at ACL ...Read more",
"title": "DeepSeek's Liang Wenfeng & Peking University's Yang ...",
"url": "https://eu.36kr.com/en/p/3401632759482502"
},
{
"snippet": "by W Che · 2025 · Cited by 3 — Senior Area Chairs, Area Chairs, reviewers, and the Best Paper Committee (led by Rada Mihalcea and Roi Reichart), whose dedication ensured the ...Read more",
"title": "ACL 2025 The 63rd Annual Meeting of the Association for ...",
"url": "https://aclanthology.org/2025.acl-long.0.pdf"
},
{
"snippet": "We're excited to share that our work on multilingual resources has won two major awards at ACL 2025: Best Resource Paper and Best SemEval Task!",
"title": "GippLab wins ACL Best Paper Awards",
"url": "https://gipplab.uni-goettingen.de/gipplab-wins-acl-best-paper-awards/"
},
{
"snippet": "Deepseek just won the best paper award at ACL 2025 with a breakthrough innovation in long context, a model using this might come soon.Read more",
"title": "Deepseek just won the best paper award at ACL 2025 with ...",
"url": "https://www.reddit.com/r/LocalLLaMA/comments/1mdn6dp/deepseek_just_won_the_best_paper_award_at_acl/"
},
{
"snippet": "Congratulations to all the outstanding and best paper award winners at #ACL2025NLP! [A few randomly selected photos from the awards session].Read more",
"title": "Congratulations to all the outstanding and best paper ...",
"url": "https://x.com/aclmeeting/status/1950745647214161930"
},
{
"snippet": "Best Paper Award at ACL 2025! Happy to share that our ACL paper, “Speed Without Sacrifice: Fine‑Tuning Language Models with Medusa and ...",
"title": "🏆 Best Paper Award at ACL 2025! 🏆 Happy to share that ...",
"url": "https://www.linkedin.com/posts/moran-beladev-762a2a176_best-paper-award-at-acl-2025-happy-activity-7356788038145863680-5gLV"
},
{
"snippet": "Record Breaking ACL 2025 Crowns Four Game-Changing Papers on Speed, Fairness & Safety for Next-Gen LLMs and Beyond | CSPaper Forum.Read more",
"title": "Record Breaking ACL 2025 Crowns Four Game-Changing ...",
"url": "https://forum.cspaper.org/topic/116/record-breaking-acl-2025-crowns-four-game-changing-papers-on-speed-fairness-safety-for-next-gen-llms-and-beyond"
},
{
"snippet": "by SH Muhammad · 2025 · Cited by 30 — In this paper, we present BRIGHTER–a collection of multi-labeled, emotion-annotated datasets in 28 different languages and across several domains.Read more",
"title": "BRIdging the Gap in Human-Annotated Textual Emotion ...",
"url": "https://aclanthology.org/2025.acl-long.436/"
},
{
"snippet": "Among the four best papers recognised by ACL, two author teams were from China. They included Liang's DeepSeek team and Yang Yaodong's team from ...Read more",
"title": "DeepSeek founder shares best paper award at top global ...",
"url": "https://www.scmp.com/tech/big-tech/article/3320255/deepseek-founder-shares-best-paper-award-top-global-ai-research-conference"
}
]
}
],
"A Theory of Response Sampling in LLMs: Part Descriptive and Part Prescriptive": [
{
"page": 1,
"results": [
{
"snippet": "by S Sivaprasad · 2024 · Cited by 1 — We study this sampling behavior and show that this underlying heuristics resembles that of human decision-making: comprising a descriptive component.Read more",
"title": "[2402.11005] A Theory of Response Sampling in LLMs",
"url": "https://arxiv.org/abs/2402.11005"
},
{
"snippet": "by S Sivaprasad · 2025 · Cited by 6 — We study this sampling behavior and show that this underlying heuristics resembles that of human decision-making: comprising a descriptive component.Read more",
"title": "A Theory of Response Sampling in LLMs: Part Descriptive ...",
"url": "https://aclanthology.org/2025.acl-long.1454/"
},
{
"snippet": "by S Sivaprasad · 2025 · Cited by 5 — When an LLM samples from multiple possibilities of a concept, the sampling heuristics is driven by a descriptive component (the statistical norm ...Read more",
"title": "A Theory of Response Sampling in LLMs: Part Descriptive ...",
"url": "https://aclanthology.org/2025.acl-long.1454.pdf"
},
{
"snippet": "by S Sivaprasad · Cited by 1 — We examine LLM response sampling and propose a theory that the sample of an LLM is driven by a descriptive component (the notion of statistical average) and ...",
"title": "Theory of LLM sampling: part descriptive and ...",
"url": "https://openreview.net/forum?id=ejvf3JrZuC"
},
{
"snippet": "Based on human cognitive studies, we propose a theory that explains the sampling heuristics to be part descriptive and part prescriptive. However, the exact ...Read more",
"title": "A Theory of Response Sampling in LLMs: Part Descriptive ...",
"url": "https://arxiv.org/html/2402.11005v4"
},
{
"snippet": "This supports the theory that an LLM's sampling behavior is influenced by a combination of descriptive norms-what is statistically likely-and ...Read more",
"title": "A Theory of Response Sampling in LLMs: Part Descriptive ...",
"url": "https://www.researchgate.net/publication/394272679_A_Theory_of_Response_Sampling_in_LLMs_Part_Descriptive_and_Part_Prescriptive"
},
{
"snippet": "By conceptualizing LLM sampling as a blend of descriptive and prescriptive influences, the theory offers a new lens for interpreting model ...",
"title": "A Theory of Response Sampling in LLMs",
"url": "https://chatpaper.com/paper/176850"
},
{
"snippet": "by S Sivaprasad · Cited by 1 — We study this heuristics and propose a theory that the sampling of an LLM is driven by a descriptive norm (the notion of statistical average) and a prescriptive ...Read more",
"title": "THEORY OF LLM SAMPLING: PART DESCRIPTIVE AND ...",
"url": "https://openreview.net/pdf/d3630fe1db2b6150ee2fe9da55ccb274cbf3c869.pdf"
},
{
"snippet": "",
"title": "PR-534: A Theory of Response Sampling in LLMs: Part ...",
"url": "https://www.youtube.com/watch?v=BTX4TO9bZ18"
},
{
"snippet": "Our paper, \"A Theory of Response Sampling in LLMs: Part Descriptive and Part Prescriptive\" has been accepted at ACL 2025 (Main Conference)!Read more",
"title": "Happy to report that our paper, \"A Theory of LLM Sampling",
"url": "https://x.com/pramod_kaushik/status/1926302792202731809"
}
]
}
],
"Fairness through Difference Awareness: Measuring Desired Group Discrimination in LLMs": [
{
"page": 1,
"results": [
{
"snippet": "by A Wang · 2025 · Cited by 13 — Fairness through Difference Awareness: Measuring Desired Group Discrimination in LLMs. Authors:Angelina Wang, Michelle Phan, Daniel E. Ho, Sanmi ...Read more",
"title": "[2502.01926] Fairness through Difference Awareness",
"url": "https://arxiv.org/abs/2502.01926"
},
{
"snippet": "by A Wang · 2025 · Cited by 13 — Fairness through Difference Awareness: Measuring Desired Group Discrimination in LLMs. In Proceedings of the 63rd Annual Meeting of the Association for ...Read more",
"title": "Fairness through Difference Awareness",
"url": "https://aclanthology.org/2025.acl-long.341/"
},
{
"snippet": "by A Wang · 2025 · Cited by 13 — Fairness through Difference Awareness: Measuring Desired Group Discrimination in LLMs. Angelina Wang. Stanford University, Cornell Tech.Read more",
"title": "Measuring Desired Group Discrimination in LLMs",
"url": "https://aclanthology.org/2025.acl-long.341.pdf"
},
{
"snippet": "Invited Talk: Fairness through Difference Awareness: Measuring Desired Group Discrimination in LLMs. Sanmi Koyejo · Angelina WangRead more",
"title": "Invited Talk: Fairness through Difference Awareness",
"url": "https://neurips.cc/virtual/2024/103988"
},
{
"snippet": "Difference-aware fairness offers an alternative perspective that explicitly acknowledges the importance of contextual differentiation. As argued ...Read more",
"title": "Measuring Desired Group Discrimination in LLMs",
"url": "https://www.researchgate.net/publication/394303187_Fairness_through_Difference_Awareness_Measuring_Desired_Group_Discrimination_in_LLMs"
},
{
"snippet": "Current generative AI models struggle to recognize when demographic distinctions matter—often leading to inaccurate and misleading outcomes.",
"title": "AI Fairness Through Difference Awareness - In Brief",
"url": "https://law.stanford.edu/stanford-lawyer/articles/ai-fairness-through-difference-awareness/"
},
{
"snippet": "This work introduces an important distinction between descriptive (fact-based), normative (value-based), and correlation (association-based) ...",
"title": "Measuring Desired Group Discrimination in LLMs",
"url": "https://www.semanticscholar.org/paper/2dfe534065f768b8b93fd40b6fac7677a9f70a05"
},
{
"snippet": "We present a benchmark suite composed of eight different scenarios for a total of 16k questions that enables us to assess difference awareness.Read more",
"title": "Measuring Desired Group Discrimination in LLMs",
"url": "https://arxiv.org/html/2502.01926v2"
},
{
"snippet": "Finally, we show results across ten models that demonstrate difference awareness is a distinct dimension to fairness where existing bias ...Read more",
"title": "Measuring Desired Group Discrimination in LLMs",
"url": "https://chatpaper.com/paper/175737"
},
{
"snippet": "The document discusses the importance of 'Difference Awareness' in algorithmic fairness, arguing that recognizing meaningful differences ...Read more",
"title": "Fairness Through Difference Awareness: Measuring ...",
"url": "https://www.scribd.com/document/905382595/Fairness-through-Difference-Awareness-Measuring-Desired-Group-Discrimination-in-LLMs"
}
]
}
],
"Language Models Resist Alignment: Evidence From Data Compression": [
{
"page": 1,
"results": [
{
"snippet": "by J Ji · 2024 · Cited by 13 — We formally deduce that fine-tuning disproportionately undermines alignment relative to pre-training, potentially by orders of magnitude.Read more",
"title": "Language Models Resist Alignment: Evidence From Data ...",
"url": "https://arxiv.org/abs/2406.06144"
},
{
"snippet": "by J Ji · 2025 · Cited by 13 — Large language models (LLMs) may exhibit unintended or undesirable behaviors. Recent works have concentrated on aligning LLMs to mitigate harmful outputs.Read more",
"title": "Language Models Resist Alignment: Evidence From Data ...",
"url": "https://aclanthology.org/2025.acl-long.1141/"
},
{
"snippet": "by J Ji · 2025 · Cited by 13 — Language models, fine-tuned with perturbations, exhibit an inverse relationship between normalized compression rate changes and dataset volume, ...Read more",
"title": "Language Models Resist Alignment: Evidence From Data ...",
"url": "https://aclanthology.org/2025.acl-long.1141.pdf"
},
{
"snippet": "by J Ji · 2024 · Cited by 8 — Language models, fine-tuned with perturbations, exhibit an inverse relationship between normalized compression rate changes and dataset volume, akin to that of ...Read more",
"title": "Language Models Resist Alignment",
"url": "https://arxiv.org/pdf/2406.06144"
},
{
"snippet": "Using compression theory, we formally derive that such fine-tuning process disproportionately undermines alignment compared to pre-training, ...Read more",
"title": "Language Models Resist Alignment",
"url": "https://openreview.net/forum?id=fuAJqdKXtQ&referrer=%5Bthe%20profile%20of%20Kaile%20Wang%5D(%2Fprofile%3Fid%3D~Kaile_Wang1)"
},
{
"snippet": "Large language models (LLMs) may exhibit unintended or undesirable behaviors. Recent works have concentrated on aligning LLMs to mitigate harmful outputs.Read more",
"title": "[ACL2025 Best Paper] Language Models Resist Alignment",
"url": "https://github.com/PKU-Alignment/llms-resist-alignment"
},
{
"snippet": "Regarding this ACL 2024 paper, this review summarizes LLM alignment resistance through compression theory, demonstrating elasticity and rebound ...Read more",
"title": "Language Models Resist Alignment: Evidence From Data ...",
"url": "https://liner.com/review/language-models-resist-alignment-evidence-from-data-compression"
},
{
"snippet": "Language Models Resist Alignment: Evidence From Data Compression. J Ji*, K Wang*, T Qiu*, B Chen*, J Zhou*, C Li, H Lou, Y Yang. ACL 2025 (Best Paper Award) ...Read more",
"title": "Tianyi Alex Qiu",
"url": "https://scholar.google.co.jp/citations?user=teiNc0sAAAAJ&hl=ja"
},
{
"snippet": "Fundamentally, our findings support a core assertion: language models exhibit elasticity, and thereby inherently resist alignment. Report issue ...Read more",
"title": "Language Models Resist Alignment: Evidence From Data ...",
"url": "https://arxiv.org/html/2406.06144v4"
},
{
"snippet": "Language Models Resist Alignment: Evidence From Data Compression ... Large language models (LLMs) may exhibit unintended or undesirable behaviors.Read more",
"title": "Language Models Resist Alignment",
"url": "https://x.com/Tianyi_Alex_Qiu/status/1950578516468105507"
}
]
}
],
"Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention": [
{
"page": 1,
"results": [
{
"snippet": "by J Yuan · 2025 · Cited by 195 — We present NSA, a Natively trainable Sparse Attention mechanism that integrates algorithmic innovations with hardware-aligned optimizations to achieve ...Read more",
"title": "Hardware-Aligned and Natively Trainable Sparse Attention",
"url": "https://arxiv.org/abs/2502.11089"
},
{
"snippet": "by J Yuan · 2025 · Cited by 195 — We present NSA, a Natively trained Sparse Attention mechanism that integrates algorithmic innovations with hardware-aligned optimizations to achieve efficient ...Read more",
"title": "Hardware-Aligned and Natively Trainable Sparse Attention",
"url": "https://aclanthology.org/2025.acl-long.1126/"
},
{
"snippet": "by J Yuan · 2025 · Cited by 195 — We present NSA, a Natively trainable Sparse Attention mecha- nism that integrates algorithmic innovations with hardware-aligned optimizations to ...Read more",
"title": "Hardware-Aligned and Natively Trainable Sparse Attention",
"url": "https://aclanthology.org/2025.acl-long.1126.pdf"
},
{
"snippet": "by J Yuan · 2025 · Cited by 195 — Sparse attention offers a promising direction for improving efficiency while maintaining model capabilities. We present NSA, a Natively ...Read more",
"title": "Hardware-Aligned and Natively Trainable Sparse Attention",
"url": "https://arxiv.org/pdf/2502.11089"
},
{
"snippet": "Efficient Triton implementations for Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention.Read more",
"title": "Efficient Triton implementations for \"Native Sparse Attention",
"url": "https://github.com/fla-org/native-sparse-attention"
},
{
"snippet": "Introducing NSA: A Hardware-Aligned and Natively Trainable Sparse Attention mechanism for ultra-fast long-context training & inference!Read more",
"title": "🚀 Introducing NSA: A Hardware-Aligned and Natively ...",
"url": "https://x.com/deepseek_ai/status/1891745487071609327?lang=en"
},
{
"snippet": "We present NSA, a Natively trainable Sparse Attention mechanism that integrates algorithmic innovations with hardware-aligned optimizations to achieve ...Read more",
"title": "Hardware-Aligned and Natively Trainable Sparse Attention ...",
"url": "https://www.reddit.com/r/MachineLearning/comments/1is9ufs/r_native_sparse_attention_hardwarealigned_and/"
},
{
"snippet": "Sparse attention offers a promising direction for improving efficiency while maintaining model capabilities. We present NSA, a Natively ...Read more",
"title": "Hardware-Aligned and Natively Trainable Sparse Attention",
"url": "https://arxiv.org/html/2502.11089v1"
}
]
}
]
},
"click": [
{
"from": "<search>ACL 2025 Best paper</search>",
"to": "https://2025.aclweb.org/program/awards/"
},
{
"from": "<search>A Theory of Response Sampling in LLMs: Part Descriptive and Part Prescriptive</search>",
"to": "https://aclanthology.org/2025.acl-long.1454/"
},
{
"from": "https://aclanthology.org/2025.acl-long.1454/",
"to": "https://aclanthology.org/2025.acl-long.1454.pdf"
},
{
"from": "<search>Fairness through Difference Awareness: Measuring Desired Group Discrimination in LLMs</search>",
"to": "https://aclanthology.org/2025.acl-long.341.pdf"
},
{
"from": "<search>Language Models Resist Alignment: Evidence From Data Compression</search>",
"to": "https://aclanthology.org/2025.acl-long.1141/"
},
{
"from": "https://aclanthology.org/2025.acl-long.1141/",
"to": "https://aclanthology.org/2025.acl-long.1141v2.pdf"
},
{
"from": "<search>Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention</search>",
"to": "https://aclanthology.org/2025.acl-long.1126/"
},
{
"from": "https://aclanthology.org/2025.acl-long.1126/",
"to": "https://aclanthology.org/2025.acl-long.1126.pdf"
}
]
}
|