ClapperText: A Benchmark for Text Recognition in Low-Resource Archival Documents
arXiv:2510.15557v1 Announce Type: cross Abstract: This paper presents ClapperText, a benchmark dataset for handwritten and printed text recognition in visually…
DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models
arXiv:2510.15015v1 Announce Type: cross Abstract: Text-to-Image (T2I) models have advanced rapidly, yet they remain vulnerable to semantic leakage, the unintended…
CoUn: Empowering Machine Unlearning via Contrastive Learning
arXiv:2509.16391v2 Announce Type: replace-cross Abstract: Machine unlearning (MU) aims to remove the influence of specific “forget” data from a trained…
Mixture of Experts Approaches in Dense Retrieval Tasks
arXiv:2510.15683v1 Announce Type: cross Abstract: Dense Retrieval Models (DRMs) are a prominent development in Information Retrieval (IR). A key challenge…
Attention-Aided MMSE for OFDM Channel Estimation: Learning Linear Filters with Attention
arXiv:2506.00452v2 Announce Type: replace-cross Abstract: In orthogonal frequency division multiplexing (OFDM), accurate channel estimation is crucial. Classical signal processing based…
LiRA: Linguistic Robust Anchoring for Cross-lingual Large Language Models
arXiv:2510.14466v1 Announce Type: cross Abstract: As large language models (LLMs) rapidly advance, performance on high-resource languages (e.g., English, Chinese) is…
From Craft to Constitution: A Governance-First Paradigm for Principled Agent Engineering
arXiv:2510.13857v1 Announce Type: cross Abstract: The advent of powerful Large Language Models (LLMs) has ushered in an “Age of the…
Does FLUX Already Know How to Perform Physically Plausible Image Composition?
arXiv:2509.21278v2 Announce Type: replace-cross Abstract: Image composition aims to seamlessly insert a user-specified object into a new scene, but existing…
STANCE: Motion Coherent Video Generation Via Sparse-to-Dense Anchored Encoding
arXiv:2510.14588v1 Announce Type: cross Abstract: Video generation has recently made striking visual progress, but maintaining coherent object motion and interactions…