Fine-Tuning with Divergent Chains of Thought Boosts Reasoning Through Self-Correction in Language Models.
Haritz PuertoTilek ChubakovXiaodan ZhuHarish Tayyar MadabushiIryna GurevychPublished in: CoRR (2024)
Keyphrases
- language model
- fine tuning
- language modeling
- viable alternative
- n gram
- speech recognition
- document retrieval
- language modelling
- information retrieval
- retrieval model
- probabilistic model
- test collection
- fine tune
- query expansion
- fine tuned
- statistical language models
- context sensitive
- ad hoc information retrieval
- query terms
- okapi bm
- retrieval effectiveness
- document ranking
- error rate
- language model for information retrieval
- language models for information retrieval
- vector space model
- word error rate
- smoothing methods
- term dependencies
- translation model
- pseudo relevance feedback
- cross lingual
- decision trees