PixelRAG shows that operating RAG entirely over web screenshots outperforms text-based retrieval on NQ, SimpleQA, MMSearch, LiveVQA, and MoNaCo, with up to 18.1% accuracy gains and 3x token savings via image compression.
hub Canonical reference
Glyph: Scaling context windows via visual-text compres- sion
Canonical reference. 83% of citing Pith papers cite this work as background.
13
Pith papers citing it
Background
83% of classified citations
hub tools
citation-role summary
background 5
baseline 1
citation-polarity summary
representative citing papers
Framing visual text compression as measure transport decomposes encoding loss into precision and coverage costs, enabling a label-free routing rule that matches oracle performance on 17 of 24 NLP datasets while using 10% fewer tokens.
TableVision benchmark shows explicit spatial grounding recovers MLLM reasoning on hierarchical tables, delivering 12.3% accuracy improvement through a decoupled perception-reasoning framework.
VISTA-Bench shows vision-language models degrade on visualized text in images compared to equivalent pure text, with larger gaps under increased perceptual difficulty.
MemShot renders local dialogue spans as structured visual memory units to improve long-term dialogue modeling in LLMs, achieving competitive benchmark performance with 70x faster memory construction.
LoMo is a lightweight data curation technique that locally substitutes text with images in prompts to enforce cross-modal invariance, yielding 2.67-2.82 point gains over standard SFT on two VLMs across 13 benchmarks.
Degraded image resolution in MLLMs bypasses safety alignments via cognitive overload, raising jailbreak rates across perturbations.
POINTS-Seeker-8B is an 8B multimodal model trained from scratch for agentic search that uses seeding and visual-space history folding to outperform prior models on six visual reasoning benchmarks.
MapTab is a new multimodal benchmark with 328 images and nearly 200k queries that shows current MLLMs have substantial difficulty with multi-criteria route planning when visual and tabular information must be combined.
Presents PopMedQA benchmark and shows domain-independent LLM methods fail on token-inefficient longitudinal medical records, leaving room for domain-specific approaches.
LensVLM trains VLMs to scan compressed rendered text images and selectively expand task-relevant regions, achieving 4.3x compression with near full-text accuracy and outperforming baselines up to 10.1x on text QA benchmarks.
MemOCR renders structured memory as images with adaptive visual density to improve long-horizon reasoning under tight context budgets.
Vision-based optical context compression performs no better than direct autoencoding baselines like mean pooling or hierarchical encoders across compression ratios.
citing papers explorer
-
PIXELRAG: Web Screenshots Beat Text for Retrieval-Augmented Generation
PixelRAG shows that operating RAG entirely over web screenshots outperforms text-based retrieval on NQ, SimpleQA, MMSearch, LiveVQA, and MoNaCo, with up to 18.1% accuracy gains and 3x token savings via image compression.
-
Visual Text Compression as Measure Transport
Framing visual text compression as measure transport decomposes encoding loss into precision and coverage costs, enabling a label-free routing rule that matches oracle performance on 17 of 24 NLP datasets while using 10% fewer tokens.
-
Memory Shot for Long-Term Dialogue
MemShot renders local dialogue spans as structured visual memory units to improve long-term dialogue modeling in LLMs, achieving competitive benchmark performance with 70x faster memory construction.
-
LoMo: Local Modality Substitution for Deeper Vision-Language Fusion
LoMo is a lightweight data curation technique that locally substitutes text with images in prompts to enforce cross-modal invariance, yielding 2.67-2.82 point gains over standard SFT on two VLMs across 13 benchmarks.
-
POINTS-Seeker: Towards Training a Multimodal Agentic Search Model from Scratch
POINTS-Seeker-8B is an 8B multimodal model trained from scratch for agentic search that uses seeding and visual-space history folding to outperform prior models on six visual reasoning benchmarks.
-
The Verbose Context Problem in Medical Records
Presents PopMedQA benchmark and shows domain-independent LLM methods fail on token-inefficient longitudinal medical records, leaving room for domain-specific approaches.
-
LensVLM: Selective Context Expansion for Compressed Visual Representation of Text
LensVLM trains VLMs to scan compressed rendered text images and selectively expand task-relevant regions, achieving 4.3x compression with near full-text accuracy and outperforming baselines up to 10.1x on text QA benchmarks.
-
MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning
MemOCR renders structured memory as images with adaptive visual density to improve long-horizon reasoning under tight context budgets.