'Mixture-of-Cross-Encoders' hypothesis fails; author eyes per-query routing
CShorten30 · x · 2026-11-16
Tweet 14/17: the author admits disappointment — the study was motivated by a 'Mixture-of-Cross-Encoders' (MoCE) idea. The models each uniquely solve dozens of queries, but a fixed blend can't turn that into better rankings, so per-query routing looks more promising. Two related papers are recommended: optimal re-ranking depth (MrParry et al.) and natural-language query-to-configuration for retrieval agents (melissapan et al.).
Related event: Reranking Study Finds Two-Stage Pipeline Beats Fusion and Deeper Windows(13 posts)→
More from Research
- BeyondBinding challenge launches: AI-designed antibodies get wet-lab tested — QuanquanGu · 2026-11-17
- Singularity-Bench: Opus 5.5 with Claude Code beats human baselines on building AI systems — armand_ruiz · 2026-11-17
- Open Vision Banana: fully open recipe turns image generators into multi-task perception systems — NielsRogge · 2026-11-17
- WebArena-Pro debuts as largest browser-agent benchmark, top agents score 62.3% — xhluca · 2026-11-17
- NeurIPS 2026 paper: RL-trained skills transfer poorly to composed reasoning tasks — StanfordAILab · 2026-11-17
- OpenAI warns CoT monitorability will degrade as models scale, researchers push back — gleech · 2026-11-17