Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
Researchers have found a way to make transfer learning methods use less memory with a new approach called "Universal Parallel Tuning" (UniPT). This makes it easier to use these advanced methods in everyday situations.
Researchers are exploring if "diffusion models" can make models that are already trained to be robust even stronger. A new method, "DiffSmooth", cleans up data using diffusion and then uses a special smoothing process to make the models even more reliable.
Langfuse is an open source observability & analytics solution for LLM-based applications. It is mostly geared towards production usage but some users also use it for local development of their LLM applications.
DSPy unifies techniques for prompting and fine-tuning LMs as well as improving them with reasoning and tool/retrieval augmentation, all expressed through a minimalistic set of Pythonic modules that compose and learn.
Sponsored
You need realistic data for demos, testing, and pre-production environments. Now you can use AI to get that data instantly, securely, and cost-effectively. Gretel's novel tabular LLM
A secure and private version of the massively popular language model is now available for enterprises. It seems to be 2x faster, with a 32k context length and no usage limits, and has significant security improvements with SOC2 compliance. An interesting statistic of the article is that 80% of all Fortune 500 companies have had employees sign up to use ChatGPT.
How many A100s does everyone have? This report likely dramatically underestimates compute, but the partial ordering seems correct. Of note - 7 published papers using Huawei's chip.
Alibaba has launched two new artificial intelligence models: Qwen-VL and Qwen-VL-Chat. The company says the models can understand images and carry out more complex conversations.
This research introduces a new method to recreate how light interacts with 3D objects, using just a few pictures of that object. Unlike older techniques, our system cleverly uses hints about shadows and highlights, allowing it to handle various shapes and materials more effectively.
This study introduces a new graph model for Multi-Camera Multi-Object Tracking (MC-MOT) that focuses on both space and time. Unlike previous methods that often make mistakes, this model first connects objects across different cameras and then uses that info over time, making tracking more accurate and ready for real-world use.
This study unveils the DFER-CLIP model, which enhances the CLIP model specifically for recognizing changing facial expressions in real-world scenarios.
An excellent model that runs quickly and generates beats, sound effects, and basic dialogue. It is easy to get running and seems to be fairly robust to prompting style.
A lightweight framework for building LLM-based agents.
Prompt2Model is a system that takes a natural language task description to train a small special-purpose model that is conducive for deployment.
An overview of the several prompting techniques that model the reasoning process as a graph structure—rather than a chain or tree—that better captures the various types of non-linear patterns that may occur when reasoning over a problem.
The excitement around AI has transitioned into a focus on practical applications, which has brought to light some challenges, particularly with LLMs as their outputs can be unpredictable. Engineers and businesses grapple with controlling chaos for structured results, while artists' hesitation to adopt AI stems from fears of job displacement and legal issues, suggesting that edgier AI models may be necessary for unlocking artistic potential.
This article examines governance strategies for advanced AI systems, focusing on evaluations, access control, and international oversight, with the goal of minimizing risk and maximizing upside.
As of this week, Microsoft is officially launching Google Chrome support for Bing AI.
DoorDash is launching a new answering service for restaurants that could put you on the phone with AI when ordering takeout or delivery.
A scalable dev tool that writes entire apps from scratch while the developer oversees the implementation.
Sponsored
Companies with stores of high quality internal data have released fine-tuned versions of Llama that are extremely adept at code just days after Meta released its new programming language models. It is important to note that the current version of GPT-4 is still better than any open-source model at coding. This article compares the new Llama models with the March version of GPT-4.
Meta AI has unveiled a series of datasets that harness the power of Unreal Engine for photorealistic imagery. The released datasets include PUG: Animals for foundation model research, PUG: ImageNet to assess image classifier robustness, and PUG: SPAR for vision-language model evaluation.
New research from Apple shows amazing performance on the challenging problem of 3D object reconstruction.
Scientists have designed a new system called MGDN that merges multiple images into one clear picture, fixing the issues older methods had in using varying images. The special tools inside MGDN help in understanding the different parts of images better and blending them smoothly.
Researchers have developed a new test called CALM to check if AI language tools are biased. After collecting a lot of data from different sources and testing various AI models, they found that some big AI models can be more biased than smaller ones.
Running at less than 1 ms latency, this super fast vision transformer from Apple is a case study in edge deployment.
While the DETR framework is popular for human-object interaction (HOI) research, current two-stage transformers overlook critical contextual details, like pose and orientation, leading to issues in identifying complex interactions. Enhanced predicate visual context (PViC) improves HOI recognition by better integrating image features and spatial information.
The reason closed source models like Claude or GPT-4 are so impressive is because they are incredibly good at reasoning AND have general purpose language ability. Many open alternatives focus on an extremely narrow niche of skills and neglect others. The OpenLemur project aims to balance language model skills across many different tasks similar to closed models.
An open source AI copilot for your SaaS product.
Sponsored
Natural language interfaces can simplify data exploration. But the data needs to be accessible and trustworthy — which is where the semantic layer comes in. Join this AtScale Tech Talk
Meta has released Code Llama, a Llama 2-based large language model for coding. The 7B, 13B, and 34B parameter models are now available. They performed strongly on HumanEval (53.7) and MBPP (56.2). Code Llama models can generate and debug code, are instruction-tuned, and have extremely long context - as much as 100k tokens.
Hugging Face has raised $235 million in a Series D funding round, with participation from Google, Amazon, Nvidia, Intel, AMD, Qualcomm, IBM, Salesforce, and Sound Ventures. The startup is now valued at $4.5 billion.
The White House, in collaboration with tech giants like Google and OpenAI, has initiated the "AI Cyber Challenge" to develop AI that can spot and fix software vulnerabilities. The competition is offering $20 million in prizes. Winning teams will be encouraged to open-source their solutions.
A pretty large and collaborative effort at constructing a benchmark for measuring language models' ability to reason in a legal framework. Watch out for contamination in already trained models.
Storytelling via automatic video generation is an emerging area of research. The tasks of story continuation, story remixing, and others are tough to measure. Hopefully, this great new Google benchmark helps the field make progress.
Most computer vision tools have trouble tracking 3D objects using LiDAR points because of distractions or not noticing long-term movements. MTM-Tracker mixes two methods and works in two stages to solve these problems.
Graphcore released an excellent Pytorch library that makes training in fp8 super easy and stable. A single model wrapper and torch compile is all you need.
Instead of using big and resource-heavy AI models, Prompt2Model lets you describe what you want in simple words and then makes a smaller, efficient AI model for that task. It can make models that work better than some popular big models.
Recent models often struggle with complex vision-language tasks due to limitations in understanding mixed image-text context. The I4 benchmark was introduced to evaluate these tasks. Results showed a flaw in the Visual Prompt Generator's attention. To fix this, researchers developed Cheetah, a model with a unique module and training strategy that excels in understanding complex interleaved instructions. Cheetah achieved top performance on the I4 benchmark.
Human feedback is key to improving language model alignment and overall performance. However, the process of training with Proximal Policy Optimization has some challenges. Recent work showed that you can optimize directly against human preference and bypass the reward model. With Text-based Reinforcement Learning, you can do this on state-of-the-art open models.
An exploration into why the current crop of LLMs are unlikely to lead to AGI and why physics itself will prevent any sudden leaps from where we are now to AGI.