Research map / Large language models / AI and society
CoChat: Enabling Bot and Human Collaboration for Task Completion
Xufang Luo, Zijia Lin, Yunhong Wang, Zaiqing Nie
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 |
|---|---|---|---|
| Train dialog manager with CoChat instead of SL only | MemHRNN · Restaurant | testing reward: 0.523 -> 0.884 | Improved |
| Train dialog manager with CoChat instead of SL only | MemHRNN · Movie | testing reward: 0.505 -> 0.841 | Improved |
| Train with CoChat instead of RL from scratch | MemHRNN · Restaurant | testing reward: 0.501 -> 0.884 | Improved |
| Train with CoChat instead of RL from scratch | MemHRNN · Movie | testing reward: 0.512 -> 0.841 | Improved |
| Train with CoChat instead of SL+RL | MemHRNN · Restaurant | testing reward: 0.791 -> 0.884 | Improved |
| Train with CoChat instead of SL+RL | MemHRNN · Movie | testing reward: 0.511 -> 0.841 | Improved |
| Replace human worker with best learned dialog manager (CoChat) | dialog manager · Restaurant | testing reward: 0.911 -> 0.884 | Hurt |
| Replace human worker with best learned dialog manager (CoChat) | dialog manager · Movie | testing reward: 0.908 -> 0.841 | Hurt |
| Online learning in CoChat instead of static SL for action suggestion | MemHRNN · Restaurant | top-5 hit rate: 82.07% -> 89.91% | Improved |
| Online learning in CoChat instead of static SL for action suggestion | MemHRNN · Movie | top-5 hit rate: 66.81% -> 82.49% | Improved |
| Online learning in CoChat instead of static SL, 3rd process | MemHRNN · Movie | top-5 hit rate: 51.14% -> 78.73% | Improved |
| Remove external memory (HRNN without memory) (ablation) | MemHRNN · Restaurant | top-5 hit rate: 90.34% -> 91.35% | Improved |
Closest papers by meaning
- CollabLLM: From Passive Responders to Active Collaborators (ICML 2025 · similarity 0.87)
- A Statistical Framework for Ranking LLM-based Chatbots (ICLR 2025 · similarity 0.84)
- Open CaptchaWorld: A Comprehensive Web-based Platform for Testing and Benchmarking Multimodal LLM Agents (NeurIPS 2025 · similarity 0.84)
- First-Person Fairness in Chatbots (ICLR 2025 · similarity 0.83)
- LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (ICLR 2025 · similarity 0.83)
- OpenChat: Advancing Open-source Language Models with Mixed-Quality Data (ICLR 2024 · 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.
