Hear2Act (Text, Explicit Feedback): leaderboard
Metric: Optimal-solution rate (%; share of dialogues ending with the first-tier option, using the final recommendation if the turn budget runs out; 480 persona-grounded consumer-service scenarios with a hidden user concern and verifiable outcomes; explicit lexical feedback: the concern is stated in the user's words; text LLM reading the dialogue transcript only, two runs per scenario). Source: arxiv.org. Saturation forecast: Around December 2026. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 74.9 |
| 2 | Kimi K2.5 | 69.4 |
| 3 | GLM-5 | 67.3 |
| 4 | Qwen 3 32B | 64.3 |
| 5 | DeepSeek V3.2 | 54.8 |
Interactive version: theaggregate.ai/benchmark?slug=hear2act-text-explicit-feedback · How It Works · Data refreshed daily, snapshot 2026-09-29.