Papersby GlyphSmolVLM: Redefining small and efficient multimodal models research paper by Hugging Face, 2025
Hugging Face · Inference and efficiency · Apr 7, 2025 · 330 citations · 212 upvotes · unverified
What it shows
SmolVLM, a series of compact multimodal models, achieves high performance with minimal GPU memory usage, making efficient deployment on mobile and edge devices possible.
By Andrés Marafioti, Orr Zohar, Miquel Farré and 14 more · arXiv 2504.05299 · PDF
UnverifiedHugging Face's summary; not yet checked by hand.
More from Hugging Face
All 10RoboColosseum: Evaluating Physical Intelligence at Scale In-The-WildRoboColosseum is an evaluation framework that tests robot AI models across different environments and physical setups.
Hugging Face50 upvotesRead →What masking geometry works best for EEG foundation models?Spatio-temporal masking strategies for EEG foundation models were systematically compared across MAE and JEPA frameworks, revealing robust performance across configurations and a novel JEPA failure mode, with optimal masking achieving high downstream accuracy at reduced compute.Hugging Face3 upvotesRead →Can We Defend Against AI-Generated Video Attacks on Real-World Crisis Events? A Systematic Evaluation of Detectors, Generators and Social DisseminationA new benchmark for AI-generated video detection reveals that current detectors fail to generalize across realistic crisis-related videos and become less reliable as content spreads socially.Hugging Face1 citations285 upvotesRead →Seeing the Needle in the Haystack: Towards Weakly-Supervised Log Instance Anomaly Localization via Counterfactual PerturbationLogMILP is a weakly supervised framework for log anomaly detection that enables both bag-level detection and instance-level localization using prototype-guided structural modeling with counterfactual perturbation consistency regularization.Hugging Face4 upvotesRead →Qualixar OS: A Universal Operating System for AI Agent OrchestrationQualixar OS enables universal AI agent orchestration through a comprehensive runtime environment supporting diverse LLM providers, agent frameworks, and communication protocols, featuring advanced multi-agent topologies, adaptive routing, and robust validation mechanisms.Hugging Face20 upvotesRead →FineVision: Open Data Is All You NeedFineVision, a large-scale and curated dataset, enhances vision-language models through rigorous data collection, de-duplication, and human oversight, leading to improved performance.Hugging Face53 citations81 upvotesRead →BLOOM: A 176B-Parameter Open-Access Multilingual Language ModelBLOOM, a 176B-parameter open-access language model, achieves competitive performance on various benchmarks and is released publicly with a Responsible AI License.Hugging Face3,165 citations41 upvotesRead →Datasets: A Community Library for Natural Language ProcessingDatasets is a community library for NLP that standardizes interfaces, versioning, and documentation for over 650 datasets with contributions from more than 250 individuals.Hugging Face828 citations16 upvotesRead →
Other inference and efficiency papers
TopicEfficient Memory Management for Large Language Model Serving with PagedAttentionvLLM manages the KV cache like pages of virtual memory, serving models with 2 to 4 times the throughput.
UC Berkeley9,965 citations72 upvotesRead →FlashAttention: Fast and Memory-Efficient Exact Attention with IO-AwarenessFlashAttention computes exact attention with far fewer GPU memory reads and writes, making long sequences faster.
Stanford University5,830 citations16 upvotesRead →Group Sequence Policy OptimizationGroup Sequence Policy Optimization (GSPO) is a reinforcement learning algorithm that improves training efficiency and performance of large language models by using sequence-level importance ratios and operations.
Alibaba (Qwen)814 citations323 upvotesRead →Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse AttentionNSA, a trainable sparse attention mechanism, enhances long-context modeling efficiency without sacrificing performance, achieving improvements in speed and accuracy over full attention models.
DeepSeek596 citations157 upvotesRead →Chronos-2: From Univariate to Universal ForecastingChronos-2, a pretrained model with a group attention mechanism, achieves state-of-the-art performance in zero-shot univariate, multivariate, and covariate-informed forecasting tasks.
Amazon276 citations28 upvotesRead →Inference-Time Scaling for Generalist Reward ModelingSelf-Principled Critique Tuning enhances pointwise generative reward modeling for large language models, improving scalability and quality compared to existing methods.DeepSeek259 citations59 upvotesRead →
Amazon276 citations28 upvotesRead →Inference-Time Scaling for Generalist Reward ModelingSelf-Principled Critique Tuning enhances pointwise generative reward modeling for large language models, improving scalability and quality compared to existing methods.About this paper
- Venue
- arXiv.org
- Citations
- 330, 47 influential · Semantic Scholar
- Upvotes
- 212 · Hugging Face
- Lab
- Hugging Face · on Companies · on Acquisitions · on Releases
Changes
Influential citations rose from 44 to 47.
Citations rose from 312 to 330.
Influential citations rose from 42 to 44.
Citations rose from 305 to 312.
Influential citations rose from 38 to 42.
Citations rose from 296 to 305.
Influential citations rose from 37 to 38.
Citations rose from 294 to 296.
A first count of 37 influential citations.
A first count of 294 citations.
Found by the weekly Hugging Face scan; not yet checked by hand.