ToolBench
PulseAugur coverage of ToolBench — every cluster mentioning ToolBench across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
Arcade.dev acquires Smithery to unify AI agent tool discovery and execution
Arcade.dev has acquired Smithery, a platform for discovering and running AI agent tools, to integrate its registry with Arcade's secure action layer. This move aims to provide enterprises with a unified solution for man…
-
New HYSET method improves LLM agent tool retrieval by evaluating tool sets holistically · 3 sources tracked
Researchers have introduced HYSET, a novel method for set-level tool retrieval designed for large language model (LLM) agents. Unlike existing approaches that evaluate tools individually or sequentially, HYSET treats th…
-
New TRACE watermark ensures LLM agent trajectory provenance
Researchers have developed TRACE, a novel two-channel watermark designed to ensure the provenance of LLM agent trajectories. This system is robust against adversaries who may attempt to rebrand or substitute agents, as …
-
AI agents use automated skill description optimization to boost routing accuracy
Researchers have developed an automated pipeline to optimize skill descriptions for enterprise AI agents, significantly reducing the engineering effort required to prevent query misrouting. This pipeline achieved an ave…
-
CoHyDE method improves LLM agent tool retrieval via co-trained encoder and rewriter
Researchers have developed CoHyDE, a novel iterative co-training method designed to enhance tool retrieval for LLM agents. This approach jointly trains a dense encoder and an LLM rewriter, addressing the vocabulary mism…
-
NaviAgent improves LLM tool orchestration with bilevel planning
Researchers have developed NaviAgent, a novel system designed to improve how large language models orchestrate the use of external tools. NaviAgent employs a bilevel architecture that separates task planning from tool e…
-
Cocreli architecture enforces preconditions for reliable instruction following
Researchers have introduced Cocoreli, a novel architecture designed to enhance the reliability of autonomous agents executing human instructions. Cocoreli addresses the issue of agents proceeding with actions despite in…
-
AgentHER framework boosts LLM agent training with failed trajectory relabeling
Researchers have developed AgentHER, a new framework designed to improve the training of LLM agents by repurposing failed trajectories. The system adapts Hindsight Experience Replay to natural language, identifying alte…