GBV-PD projects gradient-based behavior vectors of test cases and the parameter delta of an evolved model onto a shared PCA basis to estimate directional loss changes without executing the new model.
OCR Post Correction for Endangered Language Texts
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 10roles
background 1polarities
background 1representative citing papers
Injecting pre-computed layout priors from RT-DETR into VLM prompts raises markdown F1 from 0.37 to 0.92 on a 10k-page OOD benchmark and cuts infinite-loop failures across domains.
MedStruct-S benchmark shows encoder-only models outperform larger decoder-only ones on key-conditioned QA from noisy OCR clinical reports, with fine-tuned large models winning only when scale is ignored.
Spectral Tempering derives an adaptive scaling factor γ(k) from the embedding eigenspectrum via local SNR analysis and knee-point normalization to achieve near-optimal compression without training or validation.
IPO-Mine releases a toolkit and large multimodal dataset for structured analysis of IPO filings and shows state-of-the-art models diverge from human judgments on chart quality and misleadingness.
Entity signals cover only 19.7% of relevant documents on Robust04 and no configuration among 443 systems improves MAP by more than 0.05 in open-world evaluation, despite gains when entities are pre-restricted.
A new 27k-sentence Arabic-Russian parallel corpus supports fine-tuned LLM translation benchmarks that improve BLEU by 4.36 and COMET by 0.051 over zero-shot baselines for scientific content.
UniTranslator adds an Understand-Generation Alignment Module and Spatial Mask Decoder to a unified multimodal model to fix translation inconsistency and spatial misalignment in in-image machine translation, reporting SOTA results on multiple benchmarks.
Mixed training of Qwen3-Embedding-4B on legal data plus SQuAD-pt yields higher average NDCG@10 (0.447), MRR@10 (0.595), and MAP@10 (0.308) across six Portuguese retrieval datasets than legal-only or base models, with largest gains on out-of-domain question-based search.
ESG-adapted versions of Qwen-3-4B using LoRA and IRM outperform the base model and Llama-3/Gemma-3 baselines on generative ESG question-answering tasks.
citing papers explorer
-
Evolution-Aware Regression Test Prioritization of ML-Enabled Systems Using Gradient-Based Behavior Vectors
GBV-PD projects gradient-based behavior vectors of test cases and the parameter delta of an evolved model onto a shared PCA basis to estimate directional loss changes without executing the new model.
-
MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
MedStruct-S benchmark shows encoder-only models outperform larger decoder-only ones on key-conditioned QA from noisy OCR clinical reports, with fine-tuned large models winning only when scale is ignored.
-
Spectral Tempering for Embedding Compression in Dense Passage Retrieval
Spectral Tempering derives an adaptive scaling factor γ(k) from the embedding eigenspectrum via local SNR analysis and knee-point normalization to achieve near-optimal compression without training or validation.
-
IPO-Mine: A Toolkit and Dataset for Section-Structured Analysis of Long, Multimodal IPO Documents
IPO-Mine releases a toolkit and large multimodal dataset for structured analysis of IPO filings and shows state-of-the-art models diverge from human judgments on chart quality and misleadingness.
-
Bridging Scientific Heritage: An Arabic--Russian Parallel Corpus and LLM Benchmark for Sustainable Knowledge Transfer
A new 27k-sentence Arabic-Russian parallel corpus supports fine-tuned LLM translation benchmarks that improve BLEU by 4.36 and COMET by 0.051 over zero-shot baselines for scientific content.
-
UniTranslator: A Unified Multi-modal Framework for End-to-end In-Image Machine Translation
UniTranslator adds an Understand-Generation Alignment Module and Spatial Mask Decoder to a unified multimodal model to fix translation inconsistency and spatial misalignment in in-image machine translation, reporting SOTA results on multiple benchmarks.
-
Domain-Adaptive Dense Retrieval for Brazilian Legal Search
Mixed training of Qwen3-Embedding-4B on legal data plus SQuAD-pt yields higher average NDCG@10 (0.447), MRR@10 (0.595), and MAP@10 (0.308) across six Portuguese retrieval datasets than legal-only or base models, with largest gains on out-of-domain question-based search.
-
Developing an ESG-Oriented Large Language Model through ESG Practices
ESG-adapted versions of Qwen-3-4B using LoRA and IRM outperform the base model and Llama-3/Gemma-3 baselines on generative ESG question-answering tasks.