Oryx MLLM: On-Demand Spatial-Temporal Understanding at Arbitrary Resolution
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:2E3ZQERLrecord.jsonopen to challenge →
read the original abstract
Visual data comes in various forms, ranging from small icons of just a few pixels to long videos spanning hours. Existing multi-modal LLMs usually standardize these diverse visual inputs to a fixed resolution for visual encoders and yield similar numbers of tokens for LLMs. This approach is non-optimal for multimodal understanding and inefficient for processing inputs with long and short visual contents. To solve the problem, we propose Oryx, a unified multimodal architecture for the spatial-temporal understanding of images, videos, and multi-view 3D scenes. Oryx offers an on-demand solution to seamlessly and efficiently process visual inputs with arbitrary spatial sizes and temporal lengths through two core innovations: 1) a pre-trained OryxViT model that can encode images at any resolution into LLM-friendly visual representations; 2) a dynamic compressor module that supports 1x to 16x compression on visual tokens by request. These design features enable Oryx to accommodate extremely long visual contexts, such as videos, with lower resolution and high compression while maintaining high recognition precision for tasks like document understanding with native resolution and no compression. Beyond the architectural improvements, enhanced data curation and specialized training on long-context retrieval and spatial-aware data help Oryx achieve strong capabilities in image, video, and 3D multimodal understanding simultaneously. Our work is open-sourced at https://github.com/Oryx-mllm/Oryx.
This paper has not been read by Pith yet.
Forward citations
Cited by 29 Pith papers
-
ViQ: Text-Aligned Visual Quantized Representations at Any Resolution
ViQ presents a text-aligned visual quantization method with two-stage pre-training and discretization that supports native resolutions and claims competitive multimodal performance with efficiency gains.
-
VISD: Enhancing Video Reasoning via Structured Self-Distillation
VISD improves VideoLLM reasoning performance and training efficiency by combining structured multi-dimensional self-distillation feedback with RL via direction-magnitude decoupling, curriculum scheduling, and EMA stab...
-
VideoRouter: Query-Adaptive Dual Routing for Efficient Long-Video Understanding
VideoRouter uses query-adaptive semantic and image routers plus new training datasets to reduce visual tokens by up to 67.9% while improving performance over the InternVL baseline on long-video benchmarks.
-
LongVideo-R1: Smart Navigation for Low-cost Long Video Understanding
LongVideo-R1 trains a reasoning agent on 33K trajectories to intelligently select informative video clips via iterative refinement and RL, achieving better accuracy-efficiency tradeoffs on long video QA benchmarks.
-
High-Resolution Visual Reasoning via Multi-Turn Grounding-Based Reinforcement Learning
MGPO elicits grounding in LMMs via multi-turn RL with binary rewards, yielding 5.4% and 5.2% gains on MME-Realworld and V* Bench and surpassing GPT-4o on the latter after training on 21K samples.
-
ViQ: Text-Aligned Visual Quantized Representations at Any Resolution
ViQ is a new two-stage text-aligned quantization method for visual features supporting arbitrary resolutions that claims competitive multimodal performance with efficiency gains of 20-70%.
-
Don't Pause: Streaming Video-Language Synchrony for Online Video Understanding
LyraV uses FDTC and SToP for per-frame incremental decoding to reach 98.29% video synchrony at 3.89 FPS while preserving general understanding.
-
Flat-Pack Bench: Evaluating Spatio-Temporal Understanding in Large Vision-Language Models through Furniture Assembly
Flat-Pack Bench is a new evaluation suite that shows state-of-the-art LVLMs perform poorly on nuanced spatio-temporal reasoning required for furniture assembly videos.
-
See What I Mean: Aligning Vision and Language Representations for Video Fine-grained Object Understanding
SWIM aligns cross-attention maps from object nouns to ground-truth masks during training on the new NL-Refer dataset to enable text-only fine-grained video object understanding in MLLMs.
-
VideoRouter: Query-Adaptive Dual Routing for Efficient Long-Video Understanding
VideoRouter uses dual semantic and image routers for query-adaptive token compression in long-video models, delivering up to 67.9% reduction while outperforming the InternVL baseline on VideoMME, MLVU, and LongVideoBench.
-
Streaming Video Instruction Tuning
Streamo is a streaming video LLM trained end-to-end on the new Streamo-Instruct-465K dataset that unifies multiple real-time video tasks with claimed strong temporal reasoning and generalization.
-
Detector-Empowered Video Large Language Model for Efficient Spatio-Temporal Grounding
DEViL offloads spatial grounding to a detector via a distilled reference-semantic token and temporal consistency regularization, reaching 43.1% m_vIoU at 14.33 FPS on HC-STVG.
-
InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
InternVL3.5 advances open-source multimodal models with Cascade RL for +16% reasoning gains and ViR for 4x inference speedup, with the 241B model reaching SOTA among open-source MLLMs on multimodal, reasoning, and age...
-
Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligence
Spatial-MLLM boosts MLLM spatial intelligence from 2D inputs via dual encoders initialized from geometry models plus space-aware sampling, claiming state-of-the-art results.
-
Spatial-MLLM: Boosting MLLM Capabilities in Visual-based Spatial Intelligence
Spatial-MLLM adds a 3D spatial encoder initialized from a visual geometry model and space-aware frame sampling to MLLMs to improve spatial understanding and reasoning from purely 2D visual inputs.
-
VLM-3R: Vision-Language Models Augmented with Instruction-Aligned 3D Reconstruction
VLM-3R augments VLMs with implicit 3D tokens from monocular video via geometry encoding and 200K+ 3D reconstructive QA pairs, plus a new 138K-pair temporal benchmark, to support spatial and embodied reasoning.
-
InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models
InternVL3-78B sets a new open-source SOTA of 72.2 on MMMU via native joint multimodal pre-training, V2PE, MPO, and test-time scaling while remaining competitive with proprietary models.
-
MotionBench: Benchmarking and Improving Fine-grained Video Motion Understanding for Vision Language Models
MotionBench is a new benchmark showing poor fine-grained motion understanding in VLMs and proposes TE Fusion to improve performance with higher frame rates.
-
Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
InternVL 2.5 is the first open-source MLLM to surpass 70% on the MMMU benchmark via model, data, and test-time scaling, with a 3.7-point gain from chain-of-thought reasoning.
-
LiveStarPro: Proactive Streaming Video Understanding with Hierarchical Memory for Long-Horizon Streams
LiveStarPro uses SVeD for response timing via perplexity, SCAM for incremental alignment, and TSHM for event-chain memory to achieve 28.9% better semantic correctness and 1.58x speedup on long video streams.
-
VISD: Enhancing Video Reasoning via Structured Self-Distillation
VISD improves VideoLLM reasoning by adding multi-dimensional diagnostic self-distillation and RL decoupling, yielding higher accuracy, better grounding, and nearly 2x faster training convergence.
-
VISD: Enhancing Video Reasoning via Structured Self-Distillation
VISD adds structured privileged feedback from a judge model and a direction-magnitude decoupling trick to let VideoLLMs learn token-level credit assignment while keeping RL stable, yielding higher accuracy and roughly...
-
VISD: Enhancing Video Reasoning via Structured Self-Distillation
VISD proposes structured self-distillation with a multi-dimensional judge model and direction-magnitude decoupling to improve token-level credit assignment and convergence speed in VideoLLM reasoning training.
-
3D-IDE: 3D Implicit Depth Emergent
3D awareness emerges implicitly in MLLMs via self-supervised geometric constraints that create an information bottleneck, removing depth and pose dependencies at inference and cutting latency by 55%.
-
LLaVA-Octopus: Unlocking Instruction-Driven Adaptive Projector Fusion for Video Understanding
LLaVA-Octopus introduces instruction-driven adaptive fusion of multiple visual projectors in a multimodal LLM to improve video understanding performance.
-
NVILA: Efficient Frontier Visual Language Models
NVILA improves on VILA with a scale-then-compress visual token strategy and full-lifecycle efficiency optimizations, matching or exceeding leading VLMs on image and video benchmarks while reducing training cost 1.9-5....
-
LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence
LLaVA-OV-2 uses codec-stream tokenization and a shared 3D RoPE to improve video, spatial, and tracking performance over Qwen3-VL-8B, while introducing the JumpScore benchmark for fine-grained motion localization.
-
VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding
VideoLLaMA3 uses a vision-centric training paradigm and token-reduction design to reach competitive results on image and video benchmarks.
-
Toward Native Multimodal Modeling: A Roadmap
A roadmap that defines architectural nativity for multimodal models and categorizes them into Multi-to-Text, Multi-to-Target, and Multi-to-Multi types while outlining an industrial pipeline toward unified transformer-...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.