ComplLLM: Fine-tuning LLMs to Discover Complementary Signals for Decision-making
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915811973660672 |
|---|---|
| author | Guo, Ziyang Wu, Yifan Hartline, Jason Holstein, Kenneth Hullman, Jessica |
| author_facet | Guo, Ziyang Wu, Yifan Hartline, Jason Holstein, Kenneth Hullman, Jessica |
| contents | Multi-agent decision pipelines can outperform single agent workflows when complementarity holds, i.e., different agents bring unique information to the table to inform a final decision. We propose ComplLLM, a post-training framework based on decision theory that fine-tunes a decision-assistant LLM using complementary information as reward to output signals that complement existing agent decisions. We validate ComplLLM on synthetic and real-world tasks involving domain experts, demonstrating how the approach recovers known complementary information and produces plausible explanations of complementary signals to support downstream decision-makers. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_19458 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ComplLLM: Fine-tuning LLMs to Discover Complementary Signals for Decision-making Guo, Ziyang Wu, Yifan Hartline, Jason Holstein, Kenneth Hullman, Jessica Artificial Intelligence Human-Computer Interaction Multi-agent decision pipelines can outperform single agent workflows when complementarity holds, i.e., different agents bring unique information to the table to inform a final decision. We propose ComplLLM, a post-training framework based on decision theory that fine-tunes a decision-assistant LLM using complementary information as reward to output signals that complement existing agent decisions. We validate ComplLLM on synthetic and real-world tasks involving domain experts, demonstrating how the approach recovers known complementary information and produces plausible explanations of complementary signals to support downstream decision-makers. |
| title | ComplLLM: Fine-tuning LLMs to Discover Complementary Signals for Decision-making |
| topic | Artificial Intelligence Human-Computer Interaction |
| url | https://arxiv.org/abs/2602.19458 |