CompanionBench: A Theory-Anchored, Real-World-Grounded Benchmark for AI Emotional Companionship

• URL: http://arxiv.org/abs/2608.02046v2
• Date: Wed, 05 Aug 2026 09:18:25 GMT
• Title: CompanionBench: A Theory-Anchored, Real-World-Grounded Benchmark for AI Emotional Companionship
• Abstract summary: Existing benchmarks use hand-authored scenarios and prompted simulators.<n>We introduce CompanionBench, an interactive bilingual benchmark.<n>It is the first companion benchmark to ground both its scenarios and a trained user simulator in real-world data.
• Score: 7.643253506288442
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated. Existing benchmarks use hand-authored scenarios and prompted simulators, aggregate empathy into one score, and overlook judge biases such as same-family favoritism and scale drift. We introduce CompanionBench, an interactive bilingual benchmark. To our knowledge, it is the first companion benchmark to ground both its scenarios and a trained user simulator in de-identified real-world data. A hidden disclosure gate branches each persona's trajectory on the agent's own behavior, controlling the interaction state space without scripting dialogue. We operationalize ten capabilities derived from 25 theories across psychology and counseling, four of them not graded explicitly by prior work: holding ambiguity, selfobject responsiveness, positive resonance and calibrated challenge. Agents are assessed on two complementary axes: a subjective ten-capability rubric and a deterministic measure of whether deeper disclosure was earned. A cross-family panel dilutes same-family favoritism; an Item Response Theory model separates agent quality from judge severity. Theory fixes what to measure and how personas are structured; real data supply events, history, and profiles -- coverage from theory, authenticity from data. Rankings are reproducible in both languages (rho = 0.996 ZH / 0.953 EN). Evaluating 28 agents reveals capability-level differences obscured by aggregate scores. Emotion regulation and calibrated challenge remain common weaknesses; holding ambiguity discriminates most. Role-play agents rank near the bottom: immersion does not imply relational competence. Across agents, the dominant failure mode is substituting surface warmth for substantive relational support. We will release 500 Chinese-English parallel pairs and the evaluation code.

Related papers

• Disclosure-Gated User Simulation for Companion-Agent Evaluation [14.962552394114894]
  We use a large language model to play the user and train a user simulator against that specification.<n>Gating behaviour is learned from the training corpus's synthetic branch, while the real branch supplies how people speak and react.<n>We state two acceptance criteria: a ranking must be order-preserving, and absolute scores must be scale-stable.
  arXiv  Detail & Related papers  (2026-09-01T09:39:41Z)
• HarnessEval-W: Agentifying the Evaluation of Visual Worlds [149.63310396189317]
  HarnessEval-W is an agentified evaluation pipeline for benchmarking world models.<n>Rather than applying a fixed rubric, HarnessEval-W decomposes the evaluation question into measurable subproblems.<n>We apply HarnessEval-W to 18 representative world models over 330 evaluation cases.
  arXiv  Detail & Related papers  (2026-08-17T17:43:24Z)
• AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility [104.46861849039357]
  Agent systems are advancing quickly across domains, but their evaluation remains fragmented.<n>The root problem is the lack of an open, agent-agnostic assessment interface.<n>We advocate Agentified Agent Assessment (AAA), where evaluation is performed by judge agents and all participants interact through standardized protocols.
  arXiv  Detail & Related papers  (2026-06-11T17:23:54Z)
• Taming Actor-Observer Asymmetry in Agents via Dialectical Alignment [59.536125286960186]
  Multi-agent frameworks assigning specialized roles are increasingly adopted to enable self-reflection and mutual auditing.<n>We find it simultaneously induces a human-like cognitive bias known as Actor-Observer Asymmetry.<n>ReTAS guides agents to synthesize conflicting viewpoints into an objective consensus.
  arXiv  Detail & Related papers  (2026-04-21T15:05:58Z)
• Dual Optimal: Make Your LLM Peer-like with Dignity [35.58723842086038]
  Current aligned language models exhibit a dual failure mode we term the Evasive Servant.<n>We propose the Dignified Peer framework, which counters servility with anti-sycophancy and trustworthiness.
  arXiv  Detail & Related papers  (2026-04-01T14:48:44Z)
• Benchmarking at the Edge of Comprehension [38.43582342860192]
  If benchmarking becomes infeasible, our ability to measure any progress in AI is at stake.<n>We propose Critique-Resilient Benchmarking, an adversarial framework designed to compare models even when full human understanding is infeasible.<n>Unlike standard benchmarking, humans serve as bounded verifiers and focus on localized claims.
  arXiv  Detail & Related papers  (2026-02-15T20:51:29Z)
• Two-Faced Social Agents: Context Collapse in Role-Conditioned Large Language Models [0.0]
  GPT-5 exhibited complete mathematics contextual collapse and adopted a singular identity towards optimal responses.<n> Claude Sonnet 4.5 retained limited but measurable role-specific variation on the SAT items.<n>All models exhibited distinct role-conditioned affective preference, indicating that socio-affective variation can reemerge when cognitive constraints are relaxed.
  arXiv  Detail & Related papers  (2025-11-19T16:04:49Z)
• Sentient Agent as a Judge: Evaluating Higher-Order Social Cognition in Large Language Models [75.85319609088354]
  Sentient Agent as a Judge (SAGE) is an evaluation framework for large language models.<n>SAGE instantiates a Sentient Agent that simulates human-like emotional changes and inner thoughts during interaction.<n>SAGE provides a principled, scalable and interpretable tool for tracking progress toward genuinely empathetic and socially adept language agents.
  arXiv  Detail & Related papers  (2025-05-01T19:06:10Z)
• Collaborative Value Function Estimation Under Model Mismatch: A Federated Temporal Difference Analysis [55.13545823385091]
  Federated reinforcement learning (FedRL) enables collaborative learning while preserving data privacy by preventing direct data exchange between agents.<n>In real-world applications, each agent may experience slightly different transition dynamics, leading to inherent model mismatches.<n>We show that even moderate levels of information sharing significantly mitigate environment-specific errors.
  arXiv  Detail & Related papers  (2025-03-21T18:06:28Z)
• Self-Training with Pseudo-Label Scorer for Aspect Sentiment Quad Prediction [54.23208041792073]
  Aspect Sentiment Quad Prediction (ASQP) aims to predict all quads (aspect term, aspect category, opinion term, sentiment polarity) for a given review. A key challenge in the ASQP task is the scarcity of labeled data, which limits the performance of existing methods. We propose a self-training framework with a pseudo-label scorer, wherein a scorer assesses the match between reviews and their pseudo-labels.
  arXiv  Detail & Related papers  (2024-06-26T05:30:21Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.