Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Towards Awareness of Human Relational Strategies in Virtual Agents
Ian Beaver, Cynthia Freeman, Abdullah Mueen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| remove annotator-highlighted relational text from requests (ablation) | IVA input · Airline, Telecom, Train, TripAdvisor IVAs | intent confidence | Improved |
| remove highlighted relational text (A-B test of IVA responses) (ablation) | IVA input · Airline, Telecom, Train, TripAdvisor IVAs | human-judged response quality: 10 of the 16 cases | Improved |
| cleaned request changes IVA response (ablation) | IVA input · all four datasets | responses changed: 3,588 IVA responses changed out of the 27,036 | Mixed |
| remove highlighted text on TripAdvisor forum posts (agreement threshold 4) (ablation) | IVA input · TripAdvisor | A-B preference | Hurt |
| raise annotator agreement threshold from 3 to 4 (ablation) | text removal · all datasets | A-B preference | No clear effect |
| raise agreement threshold for removal (ablation) | text removal · all datasets | amount of language removed | Hurt |
| compare word-level instead of character-level alignment (ablation) | annotation alignment · all datasets | alignment score: 0.2% different | No clear effect |
| compare annotator selection agreement across domains | annotation · TripAdvisor vs Train | Fleiss kappa: 0.270 vs 0.450 | Mixed |
Closest papers by meaning
- Open Models, Closed Minds? On Agents Capabilities in Mimicking Human Personalities through Open Large Language Models (AAAI 2025 · similarity 0.86)
- Talk, Evaluate, Diagnose: User-aware Agent Evaluation with Automated Error Analysis (ICLR 2026 · similarity 0.84)
- Position: Towards Bidirectional Human-AI Alignment (NeurIPS 2025 · similarity 0.84)
- Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action (NeurIPS 2026 · similarity 0.84)
- Human Simulacra: Benchmarking the Personification of Large Language Models (ICLR 2025 · similarity 0.83)
- PieArena: Ranking and Profiling Language Agents in Realistic Negotiation Scenarios (NeurIPS 2026 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
