Pith. sign in

REVIEW 4 cited by

SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents

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

arxiv 2403.08715 v3 pith:VFMVDFTG submitted 2024-03-13 cs.CL

SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents

classification cs.CL
keywords sociallanguageagentsinteractionlearningmethodtrainingability
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Humans learn social skills through both imitation and social interaction. This social learning process is largely understudied by existing research on building language agents. Motivated by this gap, we propose an interactive learning method, SOTOPIA-$\pi$, improving the social intelligence of language agents. This method leverages behavior cloning and self-reinforcement training on filtered social interaction data according to large language model (LLM) ratings. We show that our training method allows a 7B LLM to reach the social goal completion ability of an expert model (GPT-4-based agent), while improving the safety of language agents and maintaining general QA ability on the MMLU benchmark. We also find that this training paradigm uncovers some difficulties in LLM-based evaluation of social intelligence: LLM-based evaluators overestimate the abilities of the language agents trained specifically for social interaction.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Evalet: Evaluating Large Language Models through Functional Fragmentation

    cs.HC 2025-09 conditional novelty 7.0

    Evalet applies functional fragmentation to deliver fragment-level qualitative analysis of LLM evaluations, with a user study showing 48% more misalignment detections than holistic scoring.

  2. SIV-Bench: A Video Benchmark for Social Interaction Understanding and Reasoning

    cs.CV 2025-06 conditional novelty 7.0

    SIV-Bench is a new video benchmark with 2,792 clips and 5,455 QA pairs that evaluates MLLMs on social scene understanding, state reasoning, and dynamics prediction using social relation theory.

  3. SAVOIR: Learning Social Savoir-Faire via Shapley-based Reward Attribution

    cs.AI 2026-04 unverdicted novelty 6.0

    SAVOIR combines prospective expected utility valuation with Shapley values for fair credit assignment in social dialogue RL, achieving SOTA on SOTOPIA where a 7B model matches or exceeds GPT-4o and Claude-3.5-Sonnet.

  4. DoubleAgents: Human-Agent Alignment in a Socially Embedded Workflow

    cs.HC 2025-09 unverdicted novelty 6.0

    DoubleAgents shows that a distributed-cognition design with coordination agent, dashboard, and policy module increases user comfort and reliance on AI agents for coordination tasks over time.