Tag
The paper studies faithful reasoning in AI systems for abstaining action policies, finding a tradeoff where direct policies achieve higher decision quality but lack auditable reasoning, while reasoning policies provide oversight at the cost of lower performance.
This paper evaluates the abilities of large language models (LLMs) and multimodal LLMs for addressee detection, turn-change prediction, and next speaker prediction in multi-party meeting conversations. Results show text-based LLMs outperform supervised models and humans in next speaker prediction, while multimodal LLMs improve over text-only models in other tasks but remain below human performance.
GroupMemBench is a new benchmark for evaluating LLM agent memory in multi-party conversations, exposing failures in current memory systems with the best achieving only 46% average accuracy.