DARLING, a paper fixing repetitive LLM outputs after post-training, accepted at NeurIPS
DanielKhashabi · x · 2026-09-25
The DARLING paper from Daniel Khashabi's group has been accepted to NeurIPS. N8Programs calls it one of his favorite papers and the reason he got excited about joining JHU CLSP.
The problem it tackles: post-training tends to make language models repetitive. DARLING rewards responses that are both high quality and meaningfully different from one another, using a diversity-aware reward to counter mode collapse. Worth a look for anyone tracking post-training diversity.
More from Research
- Navigating tenure-track in 2026: a guide to the academic job market — mboehme_ · 2026-09-25
- Grady Booch: LLMs Only Resemble the Brain at Its Most Primitive Structures — Grady_Booch · 2026-09-25
- ICLR 2027 Submission De-anonymization Incident Sparks OpenReview Statement — Striking-Warning9533 · 2026-09-25
- FineVision, an open-source 24.3M-sample VLM training dataset, accepted to NeurIPS — andimarafioti · 2026-09-25
- LLM Verbalized Confidence Paper Accepted at EMNLP UncertaiNLP Workshop in Budapest — yeewhye · 2026-09-25
- Valerio Capraro joins JBEE editorial board to handle AI and behavioral economics papers — ValerioCapraro · 2026-09-25