Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Augmenting Non-Collaborative Dialog Systems with Explicit Semantic and Strategic Dialog History
Yiheng Zhou, Yulia Tsvetkov, Alan W Black, Zhou Yu
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 |
|---|---|---|---|
| Add explicit FST semantic and tactic history to HED | HED · CraigslistBargain | S.acc: +23.5 | Improved |
| Add explicit FST semantic and tactic history to HED | HED · CraigslistBargain | Uni.acc: +12.7 | Improved |
| Add explicit FST semantic and tactic history to HED | HED · CraigslistBargain | Bi.acc: +8.1 | Improved |
| Add explicit FST semantic and tactic history to HED | HED · CraigslistBargain | BLEU | No clear effect |
| Add FST-DA and FST-S to model semantic and strategic history (ablation) | FeHED · CraigslistBargain | Uni.acc: 42.3 -> 49.6 | Improved |
| Add FST-DA and FST-S to model semantic and strategic history (ablation) | FeHED · Persuasion For Good | Uni.acc: 0.13 -> 0.18 | Improved |
| Add FST-S to track strategy history (ablation) | FeHED · CraigslistBargain | S.acc: 46.7 -> 61.9 | Improved |
| Add FST-DA to track dialog-act history (ablation) | FeHED · CraigslistBargain | S.acc: 49.2 -> 61.9 | Improved |
| Add strategy prediction, tracking, and decoding (ablation) | FeHED · CraigslistBargain | S.acc: 47.3 -> 61.9 | Improved |
| Replace FST history tracking with an RNN (ablation) | FeHED · CraigslistBargain | S.F1: 22.8 -> 15.5 | Hurt |
| Replace FST history tracking with an RNN (ablation) | FeHED · Persuasion For Good | Uni.acc: 0.18 -> 0.16 | Hurt |
| Add explicit FST dialog history rather than relying on vanilla HED | HED · Negotiation human evaluation, second-person | Coherent: 1.9 -> 2.8 | Improved |
Closest papers by meaning
- Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action (NeurIPS 2026 · similarity 0.86)
- DialogXpert: Driving Intelligent and Emotion-Aware Conversations Through Online Value-Based Reinforcement Learning with LLM Priors (AAAI 2026 · similarity 0.85)
- PieArena: Ranking and Profiling Language Agents in Realistic Negotiation Scenarios (NeurIPS 2026 · similarity 0.85)
- GENTEEL-NEGOTIATOR: LLM-Enhanced Mixture-of-Expert-Based Reinforcement Learning Approach for Polite Negotiation Dialogue (AAAI 2025 · similarity 0.85)
- Cooper: Coordinating Specialized Agents towards a Complex Dialogue Goal (AAAI 2024 · similarity 0.85)
- Evaluating Language Model Agency Through Negotiations (ICLR 2024 · similarity 0.85)
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.
