REVIEW 5 cited by
Making Language Models Better Tool Learners with Execution Feedback
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Tools serve as pivotal interfaces that enable humans to understand and reshape the environment. With the advent of foundation models, AI systems can utilize tools to expand their capabilities and interact with the real world. Existing tool learning methodologies, encompassing supervised fine-tuning and prompt engineering approaches, often induce large language models to utilize tools indiscriminately, as complex tasks often exceed their own competencies. However, introducing tools for simple tasks, which the models themselves can readily resolve, can inadvertently propagate errors rather than enhance performance. This leads to the research question: can we teach language models when and how to use tools? To meet this need, we propose Tool leaRning wIth exeCution fEedback (TRICE), a two-stage end-to-end framework that enables the model to continually learn through feedback derived from tool execution, thereby learning when and how to use tools effectively. Experimental results, backed by further analysis, show that TRICE can make the large language model selectively use tools by improving the accuracy of tool usage while enhancing insufficient tool learning and mitigating excessive reliance on tools. Code is available at https://github.com/zjunlp/TRICE.
Forward citations
Cited by 5 Pith papers
-
Hell or High Water: Evaluating Agentic Recovery from External Failures
Language-model agents fail badly at finding backup plans when a planned function is disabled, even when a correct alternative is guaranteed to exist.
-
PORTS: Preference-Optimized Retrievers for Tool Selection with Large Language Models
PORTS fine-tunes dense retrievers for LLM tool selection via odds-ratio preference optimization driven by a frozen LLM's log-likelihood of the gold tool call, plus contrastive semantic loss.
-
Gecko: A Simulation Environment with Stateful Feedback for Refining Agent Tool Calls
A simulated tool environment with argument validation, response synthesis, and task-state feedback improves LLM tool-call accuracy at test time.
-
O$^2$-Searcher: A Searching-based Agent Model for Open-Domain Open-Ended Question Answering
O2-Searcher uses GRPO reinforcement learning to teach a 3B LLM to search a local corpus and answer open-ended and closed-ended questions, and introduces the O2-QA benchmark.
-
SAUP: Situation Awareness Uncertainty Propagation on LLM Agent
SAUP propagates per-step uncertainty across an LLM agent's reasoning chain and weights each step by a learned situation surrogate, improving AUROC for detecting incorrect answers.
Discussion (0). Continue with ORCID to comment.