Papers with prohibitive computational cost

5 papers
Once is Enough: A Light-Weight Cross-Attention for Fast Sentence Pair Modeling (2023.emnlp-main)

Copied to clipboard

Challenge: Recent studies suggest that transformer-based models perform cross-attention over input pairs, leading to computational cost.
Approach: They propose a lightweight cross-attention mechanism that performs query encoding only once while modeling the query-candidate interaction in parallel.
Outcome: The proposed model speeds up sentence pairing by over 113x while achieving comparable performance as the more expensive models.
Training for Gibbs Sampling on Conditional Random Fields with Neural Scoring Factors (2020.emnlp-main)

Copied to clipboard

Challenge: Recent advances in NLP focus on simple approaches to model the output label space . graphical models are often limited to (heuristic) greedy search and its variants .
Approach: They propose an approach for efficiently training and decoding hybrids of graphical and graphical models based on Gibbs sampling.
Outcome: The proposed approach improves on Dutch and Dutch with graphical models . the proposed model improves over a strong baseline on three languages .
Interleaved Latent Visual Reasoning with Selective Perceptual Modeling (2026.acl-long)

Copied to clipboard

Challenge: Existing approaches to interleaved reasoning are limited by the cost of re-encoding pixel-dense images.
Approach: They propose a framework that unifies dynamic state evolution with precise perceptual modeling.
Outcome: The proposed framework outperforms existing approaches on multimodal reasoning benchmarks.
Data Efficient RLVR via Off-Policy Influence Guidance (2026.acl-long)

Copied to clipboard

Challenge: Existing data selection methods for RLVR are heuristic-based, lacking theoretical guarantees and generalizability.
Approach: They propose an off-policy influence estimation method that approximates data influence using offline trajectories.
Outcome: The proposed method reduces the computational cost of policy rollouts and improves storage and computation efficiency.
SELECting over Tokens: Curating Pre-training Data at Scale via Token Classification (2026.acl-long)

Copied to clipboard

Challenge: Existing pipelines rely on expert-crafted heuristic rules, which lack content-aware, fine-grained noise detection.
Approach: They propose a framework that reframes data refinement as a highly efficient token classification task.
Outcome: The proposed framework outperforms existing pipelines on benchmarks and is 2.5x faster at inference.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations