3 papers
cs.IR2025
Positional Bias in Long-Document Ranking: Impact, Assessment, and Mitigation
Leonid Boytsov, David Akinpelu, Nipun Katyal +5
We tested over 20 Transformer models for ranking long documents (including recent LongP models trained with FlashAttention and RankGPT models "powered" by OpenAI and Anthropic clou…
cs.LG2025
A Curious Case of Remarkable Resilience to Gradient Attacks via Fully Convolutional and Differentiable Front End with a Skip Connection
Leonid Boytsov, Ameya Joshi, Filipe Condessa
We experimented with front-end enhanced neural models where a differentiable and fully convolutional model with a skip connection is added before a frozen backbone classifier. By t…
cs.CL2025
Constrained Decoding with Speculative Lookaheads
Nishanth Nakshatri, Shamik Roy, Rajarshi Das +3
Constrained decoding with lookahead heuristics (CDLH) is a highly effective method for aligning LLM generations to human preferences. However, the extensive lookahead roll-out oper…