Inferring Latent Intentions: Attributional Natural Language Inference in LLM Agents

TL;DR

This paper introduces Attributional NLI (Att-NLI), a novel framework extending natural language inference to assess and improve LLM agents' ability to infer latent intentions in multi-agent settings, demonstrating the effectiveness of neuro-symbolic approaches.

Contribution

We propose Att-NLI, integrating social psychology principles into NLI to enable abductive and deductive inference of latent intentions in LLM agents, and validate it through a textual game experiment.

Findings

Neuro-symbolic Att-NLI agents outperform others in inference tasks.

Att-NLI enables more sophisticated intention inference in LLM agents.

Neuro-symbolic approaches achieve an average win rate of 17.08% in the game.

Abstract

Attributional inference, the ability to predict latent intentions behind observed actions, is a critical yet underexplored capability for large language models (LLMs) operating in multi-agent environments. Traditional natural language inference (NLI), in fact, fails to capture the nuanced, intention-driven reasoning essential for complex interactive systems. To address this gap, we introduce Attributional NLI (Att-NLI), a framework that extends NLI with principles from social psychology to assess an agent's capacity for abductive intentional inference (generating hypotheses about latent intentions), and subsequent deductive verification (drawing valid logical conclusions). We instantiate Att-NLI via a textual game, Undercover-V, experimenting with three types of LLM agents with varying reasoning capabilities and access to external tools: a standard NLI agent using only deductive inference, an Att-NLI agent employing abductive-deductive inference, and a neuro-symbolic Att-NLI agent performing abductive-deductive inference with external theorem provers. Extensive experiments demonstrate a clear hierarchy of attributional inference capabilities, with neuro-symbolic agents consistently outperforming others, achieving an average win rate of 17.08%. Our results underscore the role that Att-NLI can play in developing agents with sophisticated reasoning capabilities, highlighting, at the same time, the potential impact of neuro-symbolic AI in building rational LLM agents acting in multi-agent environments.