Researchers release PsyBer-Agent, a psychology-driven recommender system simulator
Researchers introduced PsyBer-Agent, a new simulator designed to improve recommender systems by modeling user dynamics like fatigue and shifting preferences. The system uses a Psy-Engine with three evolving latent states per user, calibrated against real interaction logs via Unbalanced Gromov-Wasserstein optimal transport. This avoids heavy fine-tuning of the underlying LLM while maintaining…
Key points
- PsyBer-Agent simulates user dynamics like fatigue and preference shifts via a Psy-Engine with three latent states
- Calibration uses Unbalanced Gromov-Wasserstein optimal transport to align simulated behavior with real logs
- Supports 5M concurrent tasks and outperforms prompt-only simulators in behavioral fidelity tests
The framework includes WebSim and the PsyBer Benchmark, testing seven recommender backbones across multimodal scenarios. PsyBer-Agent outperforms prompt-only simulators in behavioral fidelity and stability, handling up to 5 million concurrent simulation tasks. Code and platforms are available on 4open.science.
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- Asset learns to compress speech encoders · 1 src
- Researchers find Qwen3’s reasoning gains may stem from ‘thinking leakage’ · 1 src
- Researchers test AI agents improving LLM training strategies with up to 13.34% gains · 1 src
- Researchers model how AI and humans time new words in text · 1 src
- Researchers propose Next-Concept Modeling for faster text generation · 1 src
Comments
via GitHub Discussions