Post
2220
Test-Time Scaling for Rerankers?
Can rerankers scale at test time—not by generating longer reasoning traces, but by selectively using richer document representations?
KaLM-Reranker-V1 supports Matryoshka compression from 1× to 32×, which suggests a progressive multi-fidelity pipeline:
- Embedding retrieval → Top-100
- KaLM-Reranker @ 32× compression → Top-20
- The same reranker @ 2× compression → final ranking
The intuition is simple: cheaply screen many candidates, then allocate higher-fidelity cross-attention only to the most promising ones.
For 100@32× → 20@2×, the passage-token interaction budget is roughly 31.8% of directly running 100@2×, before fixed model overheads. The key question is whether it can retain nearly the same ranking quality.
We’re considering evaluating nDCG–latency Pareto curves.
Would you consider this a useful form of test-time scaling for retrieval?
KaLM-Embedding/KaLM-Reranker-V1-Nano
KaLM-Embedding/KaLM-Reranker-V1-Small
KaLM-Embedding/KaLM-Reranker-V1-Large
KaLM-Reranker-V1: Fast but Not Late Interaction for Compressed Document Reranking (2606.22807)
https://huggingface.co/collections/KaLM-Embedding/lychee-kalm-reranker
KaLM-Embedding
Can rerankers scale at test time—not by generating longer reasoning traces, but by selectively using richer document representations?
KaLM-Reranker-V1 supports Matryoshka compression from 1× to 32×, which suggests a progressive multi-fidelity pipeline:
- Embedding retrieval → Top-100
- KaLM-Reranker @ 32× compression → Top-20
- The same reranker @ 2× compression → final ranking
The intuition is simple: cheaply screen many candidates, then allocate higher-fidelity cross-attention only to the most promising ones.
For 100@32× → 20@2×, the passage-token interaction budget is roughly 31.8% of directly running 100@2×, before fixed model overheads. The key question is whether it can retain nearly the same ranking quality.
We’re considering evaluating nDCG–latency Pareto curves.
Would you consider this a useful form of test-time scaling for retrieval?
KaLM-Embedding/KaLM-Reranker-V1-Nano
KaLM-Embedding/KaLM-Reranker-V1-Small
KaLM-Embedding/KaLM-Reranker-V1-Large
KaLM-Reranker-V1: Fast but Not Late Interaction for Compressed Document Reranking (2606.22807)
https://huggingface.co/collections/KaLM-Embedding/lychee-kalm-reranker
