Notable data and AI papers

Papers627Labs45

627 papers · 45 labs

LiteFrame: Efficient Vision Encoders Unlock Frame Scaling in Video LLMs research paper by Google DeepMind, 2026

Google DeepMind · Inference and efficiency · May 17, 2026 · 2 citations · 26 upvotes · unverified

Read on arXiv

What it shows

LiteFrame, a lightweight video encoder with Compressed Token Distillation training method, reduces latency and increases frame processing capacity for long-form video understanding in Video LLMs while maintaining accuracy.

By Jihwan Kim, Nikhil Parthasarathy, Danfeng Qin and 5 more · arXiv 2605.17260 · PDF · Code

UnverifiedHugging Face's summary; not yet checked by hand.

More from Google DeepMind

All 12
Foundation modelsUnverifiedJul 31, 2026DiffusionGemma Technical ReportDiffusionGemma is a fine-tuned mixture-of-experts language model that uses discrete diffusion to generate text blocks in parallel, achieving high speed while preserving capabilities like multimodal inputs and reasoning.Google DeepMind14 citations40 upvotesRead →Foundation modelsJul 2, 2026Gemma 4 Technical ReportOpen multimodal models from 2.3B to 31B parameters, with a thinking mode and image and audio input.Google DeepMind356 citations81 upvotesRead →Foundation modelsUnverifiedMar 25, 2026Understanding the Challenges in Iterative Generative Optimization with LLMsGenerative optimization using large language models faces challenges due to implicit design decisions about artifact modification and learning evidence that significantly impact success across different applications.Google DeepMind11 citations27 upvotesRead →ReasoningUnverifiedMar 3, 2026LoGeR: Long-Context Geometric Reconstruction with Hybrid MemoryLoGeR enables long-term 3D video reconstruction by combining bidirectional priors with a hybrid memory system that includes parametric Test-Time Training and non-parametric sliding window attention mechanisms.Google DeepMind46 citations63 upvotesRead →Agents and evaluationUnverifiedDec 4, 2025SIMA 2: A Generalist Embodied Agent for Virtual WorldsSIMA 2, built on a Gemini foundation model, interacts in 3D virtual worlds, reasons about goals, handles complex instructions, and autonomously learns new skills through open-ended self-improvement.Google DeepMind27 citations26 upvotesRead →Multimodal and roboticsUnverifiedNov 10, 2025Robot Learning from a Physical World ModelPhysWorld integrates video generation and physical world modeling to enable accurate robotic manipulation from visual demonstrations without real robot data.Google DeepMind23 citations32 upvotesRead →Agents and evaluationUnverifiedOct 8, 2025Vibe Checker: Aligning Code Evaluation with Human PreferenceVibe Checker evaluates LLMs by combining functional correctness and instruction following to better align with human coding preferences.Google DeepMind2 citations34 upvotesRead →Multimodal and roboticsUnverifiedSep 24, 2025Video models are zero-shot learners and reasonersVeo 3, a generative video model, exhibits zero-shot capabilities across various visual tasks, suggesting a trajectory towards becoming a unified, generalist vision foundation model.Google DeepMind237 citations100 upvotesRead →

Other inference and efficiency papers

Topic
Inference and efficiencySep 12, 2023Efficient 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 →Inference and efficiencyMay 27, 2022FlashAttention: 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 →Inference and efficiencyUnverifiedJul 24, 2025Group 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 →Inference and efficiencyUnverifiedFeb 16, 2025Native 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 →Inference and efficiencyUnverifiedApr 7, 2025SmolVLM: Redefining small and efficient multimodal modelsSmolVLM, a series of compact multimodal models, achieves high performance with minimal GPU memory usage, making efficient deployment on mobile and edge devices possible.Hugging Face330 citations212 upvotesRead →Inference and efficiencyUnverifiedOct 17, 2025Chronos-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 →
About this paper
Venue
arXiv.org
Citations
2, 0 influential · Semantic Scholar
Upvotes
26 · Hugging Face
Lab
Google DeepMind · on Companies · on Acquisitions

Changes

CitationsOct 5, 2026Citations rose from 1 to 2.
CitationsSep 25, 2026A first count of 0 influential citations.
CitationsSep 25, 2026A first count of 1 citations.
New paperSep 25, 2026Found by the weekly Hugging Face scan; not yet checked by hand.