PACERec uses flow matching to adapt LLM embeddings for long-tail recommendation

_reachsumit · x · 2026-11-04

The paper proposes PACERec, an ID-free progressive adaptation framework for sequential recommendation.

• Problem: LLM item embeddings capture only static content similarity, ignore sequential transitions, and struggle to transfer interaction contexts from frequent items to long-tail items under unevenly distributed behavioral evidence
• Method: learns a shared velocity field via frequency-weighted flow matching from items with sufficient interactions, progressively guiding static LLM embeddings toward collaborative behavioral targets while preserving semantics
• Result: overcomes sparse per-item updates, generalizes sequential transition patterns to long-tail items for both user modeling and candidate scoring

Original post →

More from Research

• Noetra engineer's first project: building an image encoder from scratch for multimodal LLMs — HirokatuKataoka · 2026-11-04
• Japanese team trains image encoder from scratch to match SigLIP2, with better CJK performance — HirokatuKataoka · 2026-11-04
• SpecJudge author argues LLM-as-judge tools should never ask models for self-reported confidence — jokiruiz · 2026-11-04
• Tsinghua's BEMULA benchmark: no LLM agent memory system passes over 1.5% of forget requests — rohanpaul_ai · 2026-11-04
• KAIST AI's FactEmbed decouples query understanding to improve composed image retrieval — kaist-ai · 2026-11-04
• Intology's Locus auto-designs molecular models, beating GPT-6 and Claude agents — rohanpaul_ai · 2026-11-04

Research channel →