RefCon: Iterative Refinement and Contrastive Memory Extraction for Context-Evolving Agent
RefCon extracts agent memories via self-refinement and contrast, beating unlabeled baselines on AppWorld and BFCL-V3.
Stories per day · 30 days
2 RSS
9 stories, newest first · one card per story
Stories tagged #bfcl
AdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation
AdviSD trains a small advisor with selective self-distillation, beating advisor-GRPO on BFCL-v3 and EnvScaler.
AdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation
AdviSD trains a Qwen3-8B advisor to steer frozen Gemini and Claude executors via selective self-distillation.
WEFT: Scaling Tool-Use Post-Training for General-Purpose Agents
WEFT scales full agent systems for tool-use post-training and beats size-matched baselines on major agent benchmarks.
Imprint Reader: From Weight-Update Readout to Behavioral Intervention
An Imprint Reader describes frozen weight updates and guides edits that raise refusal and tool-use scores.
SLCA-GRPO: Resolving Cross-Segment Credit Misattribution in Tool-Calling RL
SLCA-GRPO assigns separate advantages to tool and summary tokens, improving tool-calling reinforcement learning.
Critical-State RL: Diagnosing Trainable States for Multi-Turn Tool Use
Critical-State RL identifies which multi-turn tool-use calls to train, improving missing-function task performance about 14 points on BFCL v4.
Google Research Releases ToolGrad: Answer-First Framework Hits 99.8% Pass Rate for Tool-Use Data Generation
Google Research and partners introduce ToolGrad, a verified tool-chain-first data generation framework reaching 99.8% pass rate and boosting Gemma-3-12B to 83.1 on BFCL.
10
ToolLoop: Closed-Loop Tool-Use Data Synthesis via Decomposed Generation and Dynamic Self-Feedback
ToolLoop introduces a closed-loop synthetic data framework whose 11K examples lift a 4B model to 86.40% on BFCL tool-use evaluation.