A selector trained once on LLaVA-665K in CLIP space selects 15% of instructions to reach 98.3% of full-data performance and generalizes to an unseen dataset and different VLMs.
Matthew Honnibal, Ines Montani, Sofie Van Lan- deghem, and Adriane Boyd
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4verdicts
UNVERDICTED 4representative citing papers
EMA cuts adaptation costs in learning-based systems by 14.9-42.4% and raises performance by 6.9-31.3% via state transformers for input alignment and prioritized high-utility data labeling.
UnIte selects target-domain documents for pseudo-query generation by filtering high aleatoric uncertainty and prioritizing high epistemic uncertainty, yielding +2.45 to +3.49 nDCG@10 gains on BEIR with ~4k samples.
PivotTrace selects unlabeled data for RLVR by quantifying uncertainty via pivot density from attention dynamics, outperforming full supervision using only 29.3% annotations and converging 2.75 times faster.
citing papers explorer
-
Once-For-All: A Train-Once and Select-Anytime Framework for Multimodal Instruction Tuning
A selector trained once on LLaVA-665K in CLIP space selects 15% of instructions to reach 98.3% of full-data performance and generalizes to an unseen dataset and different VLMs.
-
EMA: Efficient Model Adaptation for Learning-based Systems
EMA cuts adaptation costs in learning-based systems by 14.9-42.4% and raises performance by 6.9-31.3% via state transformers for input alignment and prioritized high-utility data labeling.
-
UnIte: Uncertainty-based Iterative Document Sampling for Domain Adaptation in Information Retrieval
UnIte selects target-domain documents for pseudo-query generation by filtering high aleatoric uncertainty and prioritizing high epistemic uncertainty, yielding +2.45 to +3.49 nDCG@10 gains on BEIR with ~4k samples.
-
Smart Picks in the Dark: Towards Efficient RLVR for Reasoning via Tracing Metacognitive Pivots
PivotTrace selects unlabeled data for RLVR by quantifying uncertainty via pivot density from attention dynamics, outperforming full supervision using only 29.3% annotations and converging 2.75 times faster.