Skip to content
Papers.

DUET-VLM: Dual stage Unified Efficient Token reduction for VLM Training and Inference research paper by AMD, 2026

AMD · Feb 21, 2026 · Inference and efficiency · 0 citations · 5 upvotes · unverified 7 months ago

Read on arXiv

What it shows

DUET-VLM presents a dual compression framework that reduces visual tokens while maintaining high accuracy in vision-language models through coordinated vision and language backbone processing.

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

More from AMD

All 6
PaperCitations
Instella-MoE Technical ReportInstella-MoE is an open Mixture-of-Experts language model trained on AMD GPUs using sparse activation, Gated Multi-head Latent Attention, and multi-stage post-training to achieve strong benchmark performance with full reproducibility.Foundation models · Sep 2026 · Unverified3 weeks ago0
Stabilizing Efficient Reasoning with Step-Level Advantage SelectionShort-context post-training induces reasoning compression but causes instability; Step-level Advantage Selection addresses this by selectively adjusting reasoning steps based on confidence and verification outcomes, improving accuracy-efficiency trade-off in reasoning tasks.Reasoning · Apr 2026 · Unverified5 months ago1
Dynamic Chunking Diffusion TransformerDynamic Chunking Diffusion Transformer adapts token sequence length based on image content and diffusion timestep, improving efficiency and performance over fixed-token approaches.Architectures · Mar 2026 · Unverified6 months ago2
Instella: Fully Open Language Models with Stellar PerformanceInstella, a family of fully open large language models, achieves state-of-the-art performance using open data and is competitive with leading open-weight models, with specialized variants for long context and mathematical reasoning.Alignment and safety · Nov 2025 · Unverified10 months ago6
XModBench: Benchmarking Cross-Modal Capabilities and Consistency in Omni-Language ModelsXModBench evaluates cross-modal consistency in OLLMs, revealing challenges in spatial and temporal reasoning, modality disparities, and directional imbalance.Agents and evaluation · Oct 2025 · Unverified11 months ago4

Other inference and efficiency papers

Topic
PaperCitations
Efficient 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 Berkeley · Sep 20233 years ago8,538
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 University · May 20224 years ago5,355
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) · Jul 2025 · Unverified1 year ago694
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.DeepSeek · Feb 2025 · Unverified1 year ago508
SmolVLM: 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 Face · Apr 2025 · Unverified1 year ago296
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.DeepSeek · Apr 2025 · Unverified1 year ago249
About this paper
Authors
Aditya Kumar Singh, Hitesh Kandala, Pratik Prabhanjan Brahma and 2 more
arXiv
2602.18846 · PDF
Venue
arXiv.org
Citations
0, 0 influential · Semantic Scholar
Upvotes
5 · Hugging Face
Code
github.com/AMD-AGI/DUET-VLM
Lab
AMD · on Companies · on Acquisitions · on Quarterly · on Paydays · on TechConf

Changes

What changed
Influential citationsfirst count: 0Sep 28, 2026today
Citationsfirst count: 0Sep 28, 2026today
New paperFound by the weekly scan, unverifiedSep 28, 2026today

New papers by email

Monday afternoons, only in weeks with new papers from the labs.

Double opt-in. Unsubscribe any time.