Self-training LLMs to use search tools is much older than 2021, researcher notes with 2020 work
lateinteraction · x · 2026-10-14
Responding to YouJiacheng in a thread on self-training LLMs to use tools — search in particular — lateinteraction (Ben Clavié) points out this line of work is much older in the literature than 2021, citing his own work from fall 2020. The exchange clarifies a common misconception: training models to invoke retrieval/search is not a recent idea, with research roots going back well before.
More from Research
- Six training strategies compared: can LLMs reason over graphs at scale? — burkov · 2026-10-14
- VisInteract: NeurIPS Spotlight reframes text-to-visualization as multi-turn intent retrieval — jiqizhixin · 2026-10-14
- WorldRecon reconstructs large-scale 3D scenes from unposed multi-view images, cutting Chamfer distance 43% — ZhejiangUniversity · 2026-10-14
- RSI-Nav: recursive self-improvement for navigation agents hits 68.6% SR on R2R-CE — OpenGVLab · 2026-10-14
- NUS proposes Jeva, a decision-native architecture with near-perfect null detection at thousands of candidates — NationalUniversityofSingapore · 2026-10-14
- RUC's Agent-Reflex trains reflective agents in three stages, scoring 58.6 on long-horizon tool-use benchmarks — RUC-NLPIR · 2026-10-14