Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Customized Nonlinear Bandits for Online Response Selection in Neural Conversation Models
Bing Liu, Tong Yu, Ian Lane, Ole Mengshoel
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 |
|---|---|---|---|
| Use nonlinear polynomial-feature reward modeling instead of linear reward modeling with the same LSTM encoding | RNN-NonLinear-TS · Ubuntu Dialogue Corpus, 1 in 10 R@1 | Recall@1: 10.5 % -> 24.5 % | Improved |
| Use nonlinear polynomial-feature reward modeling instead of linear reward modeling with the same LSTM encoding | RNN-NonLinear-TS · Ubuntu Dialogue Corpus, 1 in 10 R@2 | Recall@2: 21.0 % -> 38.0 % | Improved |
| Use nonlinear polynomial-feature reward modeling instead of linear reward modeling with the same LSTM encoding | RNN-NonLinear-TS · Ubuntu Dialogue Corpus, 1 in 10 R@5 | Recall@5: 67.5 % -> 75.0 % | Improved |
| Use LSTM text encoding instead of TF-IDF features with linear reward modeling | RNN-Linear-TS · Ubuntu Dialogue Corpus, 1 in 10 R@1 | Recall@1: 10.0 % -> 10.5 % | Improved |
| Use LSTM text encoding instead of TF-IDF features with linear reward modeling | RNN-Linear-TS · Ubuntu Dialogue Corpus, 1 in 10 R@2 | Recall@2: 19.5 % -> 21.0 % | Improved |
| Use LSTM text encoding instead of TF-IDF features with linear reward modeling | RNN-Linear-TS · Ubuntu Dialogue Corpus, 1 in 10 R@5 | Recall@5: 60.5 % -> 67.5 % | Improved |
| Use nonlinear reward modeling instead of linear reward modeling with the same LSTM encoding | RNN-NonLinear-TS · Ubuntu Dialogue Corpus, 1000 online samples, k=1 | average cumulative regret | Improved |
Closest papers by meaning
- A Multi-Agent Conversational Bandit Approach to Online Evaluation and Selection of User-Aligned LLM Responses (AAAI 2026 · similarity 0.90)
- Do Response Selection Models Really Know What’s Next? Utterance Manipulation Strategies for Multi-turn Response Selection (AAAI 2021 · similarity 0.88)
- Online Multi-LLM Selection via Contextual Bandits Under Unstructured Context Evolution (AAAI 2026 · similarity 0.88)
- Efficient Explorative Key-Term Selection Strategies for Conversational Contextual Bandits (AAAI 2023 · similarity 0.88)
- Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training (ICLR 2025 · similarity 0.85)
- Offline RL by Reward-Weighted Fine-Tuning for Conversation Optimization (NeurIPS 2025 · similarity 0.84)
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.
