Agent Q integrates MCTS-guided search, self-critique, and off-policy DPO to train LLM agents that outperform behavior cloning and reinforced fine-tuning baselines in WebShop and achieve up to 95.4% success in real-world booking scenarios.
Cross-Task Generalization via Natural Language Crowdsourcing Instructions
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
other 1
citation-polarity summary
representative citing papers
Continued pretraining of Code Llama on Proof-Pile-2 yields Llemma, an open math-specialized LLM that beats known open base models on MATH and supports tool use plus formal proving out of the box.
Auto-CoT automatically builds chain-of-thought demonstrations by sampling diverse questions and letting the LLM generate reasoning chains, matching manual CoT performance on ten reasoning tasks with GPT-3.
A lifecycle-based survey of LLM fine-tuning security that reviews attacks and defenses by intervention phase and reports unified empirical findings on model-dependent attack effectiveness and limited defense generalization.
A literature review concludes that pursuing consensus in data annotation creates biased AI by dismissing subjective disagreements and enforcing geographic hegemony, and proposes mapping diversity instead.
POVID generates AI-created preference data to fine-tune vision-language models with DPO, reducing hallucinations and improving benchmark scores.
Model developers must address human concerns, preferences, values, and goals with rigor at every stage of the LLM pipeline rather than only in post-training.
citing papers explorer
-
Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents
Agent Q integrates MCTS-guided search, self-critique, and off-policy DPO to train LLM agents that outperform behavior cloning and reinforced fine-tuning baselines in WebShop and achieve up to 95.4% success in real-world booking scenarios.
-
Llemma: An Open Language Model For Mathematics
Continued pretraining of Code Llama on Proof-Pile-2 yields Llemma, an open math-specialized LLM that beats known open base models on MATH and supports tool use plus formal proving out of the box.
-
Automatic Chain of Thought Prompting in Large Language Models
Auto-CoT automatically builds chain-of-thought demonstrations by sampling diverse questions and letting the LLM generate reasoning chains, matching manual CoT performance on ten reasoning tasks with GPT-3.
-
Security in the Fine-Tuning Lifecycle of Large Language Models: Threats, Defenses,Evaluation, and Future Directions
A lifecycle-based survey of LLM fine-tuning security that reviews attacks and defenses by intervention phase and reports unified empirical findings on model-dependent attack effectiveness and limited defense generalization.
-
The Consensus Trap: Dissecting Subjectivity and the "Ground Truth" Illusion in Data Annotation
A literature review concludes that pursuing consensus in data annotation creates biased AI by dismissing subjective disagreements and enforcing geographic hegemony, and proposes mapping diversity instead.
-
Aligning Modalities in Vision Large Language Models via Preference Fine-tuning
POVID generates AI-created preference data to fine-tune vision-language models with DPO, reducing hallucinations and improving benchmark scores.
-
Reflections and New Directions for Human-Centered Large Language Models
Model developers must address human concerns, preferences, values, and goals with rigor at every stage of the LLM pipeline rather than only in post-training.
- iPOE: Interpretable Prompt Optimization via Explanations