Tencent uses dual-path decoding to reduce structural drift in recommendations
_reachsumit · x · 2026-07-24
Tencent adapts autoregressive generation to keep recommendation structure intact
Tencent presents a recommendation method that tries to close the structural gap in autoregressive generation.
- The system restores item-level structure during encoding.
- It suppresses semantic drift with hierarchical decoding.
- The method also uses path reranking and dual-path decoding to improve recommendation quality.
More from Research
- CORTEX turns local models into an interpretability desktop with logits, attention, and interventions — JayB_Official · 2026-07-24
- IBM Research features work on standardizing AI evals — evijit · 2026-07-24
- Why agent stacks can keep taking valid steps while losing the verified state — Present-Quantity-813 · 2026-07-24
- AREX introduces a recursively self-improving deep research agent with inner and outer loops — _reachsumit · 2026-07-24
- NeurIPS reviewer joke meets Pangram AI’s all-or-nothing scoring screenshot — torchcompiled · 2026-07-24
- Editable text user profiles make recommendations more controllable — _reachsumit · 2026-07-24