2024-12-02

Models

• Site Unreachable
• voyage-code-3: more accurate code retrieval with lower dimensional, quantized embeddings – Voyage AI
• Introducing Moondream 0.5B: The World’s Smallest Vision-Language Model

Papers

• [2411.18363] ChatRex: Taming Multimodal LLM for Joint Perception and Understanding
• [2411.18207v1] From Open Vocabulary to Open World: Teaching Vision Language Models to Detect Novel Objects
• [2411.18674v1] Active Data Curation Effectively Distills Large-Scale Multimodal Models #star
• [2411.19865] Reverse Thinking Makes LLMs Stronger Reasoners
• [2411.18933v1] Efficient Track Anything
• [2411.16085] Cautious Optimizers: Improving Training with One Line of Code
• [2411.16828] CLIPS: An Enhanced CLIP Framework for Learning with Synthetic Captions
• [2412.00714] Scaling New Frontiers: Insights into Large Recommendation Models
• [2411.16205] MH-MoE: Multi-Head Mixture-of-Experts
• [2412.01720] LamRA: Large Multimodal Model as Your Advanced Retrieval Assistant
• [2412.00965v1] Token Cropr: Faster ViTs for Quite a Few Tasks
• [2412.01818] [CLS] Attention is All You Need for Training-Free Visual Token Pruning: Make VLM Inference Faster
• [2412.01822v1] VLsI: Verbalized Layers-to-Interactions from Large to Small Vision Language Models
• [2411.19842v1] Scaling Transformers for Low-Bitrate High-Quality Speech Coding
• [2412.01562] Detection, Pose Estimation and Segmentation for Multiple Bodies: Closing the Virtuous Circle
• [2410.23970] TrAct: Making First-layer Pre-Activations Trainable
• [2412.01824] X-Prompt: Towards Universal In-Context Image Generation in Auto-Regressive Vision Language Foundation Models
• [2412.01819] Switti: Designing Scale-Wise Transformers for Text-to-Image Synthesis
• [2412.01940] Down with the Hierarchy: The ‘H’ in HNSW Stands for “Hubs”
• [2412.01951] Self-Improvement in Language Models: The Sharpening Mechanism
• [2210.13452] MetaFormer Baselines for Vision
• [2412.03555] PaliGemma 2: A Family of Versatile VLMs for Transfer
• Probabilistic weather forecasting with machine learning | Nature
• [2412.03561v1] FLAIR: VLM with Fine-grained Language-informed Image Representations
• [2411.16318] One Diffusion to Generate Them All
• [2412.04431] Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis
• [2412.04468] NVILA: Efficient Frontier Visual Language Models
• One Step Diffusion via Shortcut Models | OpenReview
• InternVL/InternVL2_5_report.pdf at main · OpenGVLab/InternVL · GitHub
• [2412.04234v1] DEIM: DETR with Improved Matching for Fast Convergence
• [2412.04332] Liquid: Language Models are Scalable Multi-modal Generators
• [2412.03215] Beyond [cls]: Exploring the true potential of Masked Image Modeling representations

Code

• GitHub - IDEA-Research/ChatRex: Code for ChatRex: Taming Multimodal LLM for Joint Perception and Understanding
• GitHub - 343gltysprk/ovow
• GitHub - Tencent/HunyuanVideo: HunyuanVideo: A Systematic Framework For Large Video Generation Model Training
• GitHub - benbergner/cropr
• GitHub - Theia-4869/FasterVLM: Official code for paper: [CLS] Attention is All You Need for Training-Free Visual Token Pruning: Make VLM Inference Faster.
• GitHub - fishaudio/fish-speech: SOTA Open Source TTS

Articles

• Diffusion Meets Flow Matching
• Optimizing ColPali for Retrieval at Scale, 13x Faster Results - Qdrant
• OpenAI’s o1 using “search” was a PSYOP - by Nathan Lambert
• GenCast predicts weather and the risks of extreme conditions with state-of-the-art accuracy - Google DeepMind

Videos

• The Hitchhiker’s Guide to Reasoning - YouTube #star
• Aaron Defazio Talk (12.06.2024, UCLA) - YouTube
• NEURAL NETWORKS ARE REALLY WEIRD… - YouTube

Other

• GitHub - huggingface/smol-course: A course on aligning smol models.
• arsaporta/symile-m3 · Datasets at Hugging Face

Tweets