Alibaba paper calibrates document-local boundaries to improve contextual chunk retrieval
_reachsumit · x · 2026-11-11
An Alibaba team's arXiv paper (2611.12097) tackles Late Chunking retrieval: shared context makes semantically similar same-document negatives hard to distinguish. Building on InSeNT, they propose a two-part framework—an adaptive discriminative scale from within-document margins and chunk-aware attention during document encoding. Experiments on ConTEB and NanoBEIR show improved contextual chunk retrieval across encoder settings while keeping competitive passage-level quality.
More from Research
- RoboScript hits 83.4% success on RoboTwin 2.0 with one demo per task, beating trained policies by 20+ points — H-EmbodVis · 2026-11-11
- Alibaba's PhysOPSD uses executable code as privileged context to make video generation physically plausible — AlibabaTongyiLab · 2026-11-11
- mayfer: clean low-rank latents may be the key to continuous online learning — mayfer · 2026-11-11
- PhD student finds lost Mayan city with 6,000+ buildings in old LiDAR dataset — bilawalsidhu · 2026-11-11
- Scaling intelligence: pattern-discovery problems all benefit from test-time scaling — QuanquanGu · 2026-11-11
- OXMAN hiring Systems Architect for EDEN, a generative ecosystem digital twin — tobias_rees · 2026-11-11