PrefixMem encoder for Semantic IDs improves deepest-level accuracy by up to 46% relative and full-SID retrieval recall by up to 22% relative on Pinterest data across LLM families.
Vectorizing the trie: Efficient constrained decoding for llm-based generative retrieval on accelerators
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.IR 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
UniPinRec unifies retrieval and ranking into a single model and pipeline deployed at Pinterest, reporting +1% engagement lift, 11.1% lower latency, and 63.6% higher QPS.
CapsID uses probabilistic capsule routing and confidence-based termination to generate variable-length semantic IDs, improving recall by 9.6% over strong baselines with half the latency of dual-representation systems.
citing papers explorer
-
LLMs Need Encoders for Semantic IDs Too
PrefixMem encoder for Semantic IDs improves deepest-level accuracy by up to 46% relative and full-SID retrieval recall by up to 22% relative on Pinterest data across LLM families.
-
UniPinRec: Unifying Generative Retrieval and Ranking at Pinterest Scale
UniPinRec unifies retrieval and ranking into a single model and pipeline deployed at Pinterest, reporting +1% engagement lift, 11.1% lower latency, and 63.6% higher QPS.
-
CapsID: Soft-Routed Variable-Length Semantic IDs for Generative Recommendation
CapsID uses probabilistic capsule routing and confidence-based termination to generate variable-length semantic IDs, improving recall by 9.6% over strong baselines with half the latency of dual-representation systems.