QA Wolf has multiple case studies of customers saving $200k+/year
in QA engineering and infrastructure costs.
Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
in QA engineering and infrastructure costs.
WhisperKit integrates OpenAI's Whisper speech recognition model with Apple's CoreML framework for efficient local inference on Apple devices with Swift. It can be integrated with Swift projects using the Swift Package Manager. WhisperKit automatically downloads recommended models, but it can also create and deploy fine-tuned versions. A link to a demo app is available in the repository.
Performance is a deep and nuanced domain. There's a lot that can go wrong and how sites manage resources after-load can have huge impacts on perceived performance. Global baselines matter as many teams have low performance management maturity and many popular frameworks fail to ward against catastrophic results. This article looks at current web technologies to provide an up-to-date understanding for working web developers. It looks at what's changed since last year and what technologies new projects should consider.
There were no public text-to-video models available at the beginning of 2023, but now there are dozens in active use with millions of users worldwide. While many of these products are still limited, the progress in the past year suggests that the industry is about to massively transform. This article looks at the biggest developments so far in AI video generation, companies to keep an eye on, and the remaining underlying questions in the space.
ChatGPT can assist people during their technical interviews. An experiment on how much the tool can help and how easy it is to cheat uncovered an urgent need for companies to immediately change the types of interview questions they're asking. Interviewees in the study were able to trick professional interviewers and pass interviews without the interviewers suspecting them of cheating. Not only were interviewees not caught cheating, the interviewers also rated their confidence in their hiring decisions as high.
Sponsored
Tired of unproductive meetings? Join Shopify, Uber, and Gong using Fellow [15], G2's #1 Meeting Productivity tool. Features AI meeting recording, transcription, summaries, collaborative agendas. Save time, enhance collaboration.
Crypto and AI are two major technology trends from the last decade - while the technologies can be combined, there aren't that many useful applications where the two fields intersect.
Layer-Selective Rank Reduction (LASER) is a technique that involves replacing one weight matrix with an approximate smaller one, making it so the model relies less on it - the technique makes large language models more accurate.
Starlink passes terabits per second of data every day across 9,000 lasers that can sustain a 100Gbps connection per link.
Infinite Craft is a game where you combine two elements together to create new elements, which you can then use to combine with other elements to create new elements, which you can then use to combine with other elements to...
Benchmarks are useful tools, but they're not the best indicators of real-world utility due to their narrow scopes, overfitting, contamination, reproducibility issues, and lack of scope.
A discussion on change by Vitalik Buterin, one of Ethereum's co-founders.
Sponsored
Building production-ready AI applications is hard and expensive. Anyscale [5] empowers builders to iterate fast with a simple user interface. Fine-tuning and deployment are 10x faster with automatic scaling and proactive resource management. You get the best cost-performance benefits using serverless or fully hosted on Ray. Ray is trusted by OpenAI, Uber, AWS, Spotify [5] and other AI leaders. Now, you can build your GenAI, conversational AI, or computer vision app easily — here’s how:
The AI community is abuzz with a leaked large language model called "miqu-1-70b" that was shared on HuggingFace and 4chan. The model shows performance rivaling OpenAI's GPT-4 and is linked to Mistral's Mixtral 8x7b. This incident, acknowledged by Mistral's CEO as a leak by an enthusiastic customer, highlights the growing capabilities of open source AI.
Dipendra Misra presented Layer-Selective Rank Reduction (LASER), a technique that improves large language model (LLM) accuracy by replacing a weight matrix with a smaller approximation, at the January Microsoft Research Forum.
The newest version of Llava, a vision language model, has improved OCR, reasoning, and world knowledge. It matches Gemini on some tasks. The Llava team plans to release data, code, and the model.
Weaver is a series of models specifically trained to tell stories. The largest model (34B params) outperforms GPT-4 on a storytelling benchmark.
This paper introduces a benchmark and two datasets for scene and handwritten texts. Featuring original, corrupted, and assistant images, the novel Global Structure-guided Diffusion Model (GSDM) leverages text structure to efficiently recover clean texts. It shows significant improvements in recognition accuracy and image quality.
Norton is a new approach to improving long-term video understanding by addressing the multi-granularity noisy correspondence issue in video-language studies.
MF-MOS is a new model for better detecting moving objects in autonomous driving using LiDAR technology. It uniquely separates motion and semantic information, using residual maps for motion capture and range images for semantic guidance.
Monte Carlo Tree Search (MCTS) is a state-of-the-art search algorithm. It was used in AlphaGo and researchers are now working to integrate it with language models. MCTS is notoriously tricky to implement.
AI has gotten good enough to accomplish multiple tasks in under a minute, which means organizations and employees are going to have to emphasize using AI for good, not bad.
FireLlava is a new open vision model trained on data that makes it possible to use for commercial tasks. It matches the performance of the original Llava, but not yet to the level of Llava 1.5.
ServiceNow, a $150b market cap company, last week announced that its gen AI products drove the largest net new ACV contribution for its first full quarter of any of its new product family releases ever, including its original Pro SKU. We’re already seeing enterprise-level AI applications drive meaningful revenue growth – that’s cool.
Sponsored
Fly to Hawaii for $149, Greece for $289, or anywhere else. Get 93%
before midnight. Comes with 3-day money-back guarantee*
A library that allows for creating autonomous AI agents in Python.
GPAvatar introduces a new method for creating 3D head avatars from images that use a dynamic point-based expression field and a Multi Tri-planes Attention module.
Automate resume screening with AI.
PS. Ref Finance's Meme Season [6] is live now!
Sponsored
Experience the full power and flexibility of the NEAR protocol with Ref Finance [5]. Trade NEAR, STNEAR, DAI, USDt, and many other tokens [5]. Stake REF tokens to earn more.
The estate of bankrupt crypto exchange FTX announced that it believes it can fully repay customers whose funds were caught up in the implosion of the exchange in 2022. This repayment option emerged as the best path forward after plans to relaunch the platform were abandoned due to a lack of interested buyers. The news brings hope to thousands of investors who filed claims, with the total reportedly reaching $16 billion.
Jupiter successfully distributed about $700m of its JUP token to nearly a million wallets in one of the largest token airdrops in recent history. The JUP token quickly shot up to $0.72 after launching at $0.41. Despite a heavy load on the chain, Solana's network remained stable, defying the expectations of some skeptics. Some users experienced difficulties in the early stages of the airdrop due to overwhelmed RPC nodes.
Move Labs has launched M2, the first-ever Move-EVM L2 on Ethereum that offers parallelization and formal verification. Using Block-STM parallelization technology from Diomede, M2 optimizes transaction speed and enables a variety of Move and Ethereum EVM smart contracts. By managing the execution environment separately from any consensus mechanism, M2 enhances security, adaptability, and scalability.
Ethereum's biggest upgrade since 2023, Dencun, which will roll out proto-danksharding, is closer to becoming a reality after successfully going live on the second of three test network phases. The upgrade is expected to reduce transaction costs for L2s and make data availability cheaper. Dencun is scheduled to go live on its final testnet on February 7, after which a date for the main blockchain activation will be decided upon.
Berachain might be a best-case scenario for users due to its unique value flow mechanism. Unlike traditional PoS chains, Berachain's design offers economic rewards for liquidity providers through trading fees and BGT rewards. Delegation of BGT to a validator can lead to earning block rewards and validator bribes. The design is ultimately aimed at ensuring most profits from validators and protocols are returned to its user base.
This thread discusses the pros and cons of airdrop farming. While some crypto users complain about airdrop distributions being flawed and diluted, others are making significant returns by farming these drops. The author advocates for airdrop farming, which encourages people to try out apps, suggesting that it might continue to be lucrative and help users come up with better solutions or designs for allocating tokens.
In this post, Vitalik Buterin, co-founder of Ethereum, reflects on the fast-paced evolution of the crypto space, the shift of his role from wunderkind to a different position, and the necessity of new generations taking the lead. He discusses the need for the crypto community to remind people of the relevance of money and finance and to build a broader, alternative tech stack. Buterin highlights the symbiotic relationship between token holders, intellectuals, builders, and users in the crypto ecosystem. He also explores the global aspect of crypto, referencing experiences in places like China, Silicon Valley, Taiwan, and Argentina.
Hong Kong's privacy watchdog, PCPD, has launched an investigation into iris-scanning cryptocurrency firm Worldcoin’s potential serious risks to personal data privacy. The probe will ascertain if Worldcoin has breached the Personal Data Privacy Ordinance. PCPD advises the public to consider the legitimacy and purpose of data collection before handing over biometric data. Worldcoin, whose token value recently dropped by more than 10%, rewards people with tokens in return for iris scans to prove their identity.
Binance's GitHub data, including passwords and site code, were leaked recently, posing a severe financial and security risk. It was removed without evidence of misuse.
A Ripple Co-Founder’s personal XRP account was breached yesterday. He claims the team quickly resolved the issue, notified exchanges to freeze affected addresses, and involved law enforcement.
Tether's Q4 2023 report revealed a $2.85B profit, with a total annual profit of $6.2B. The company continued to use some of these proceeds to acquire Bitcoin and Gold to put on its balance sheet.
Craig Wright's litigation against those denying his claim to be Satoshi Nakamoto has prompted the Crypto Open Patent Alliance to file a case to legally refute his claims, highlighting Wright's misuse of the legal system to intimidate and harm the crypto community.
In the fallout of a deal that saw the company placed under the microscope of competition authorities in Europe and the US, Adobe has confirmed it has “no plans to further invest” in its UI/UX web design software. Support will continue for existing users. Adobe remains open to new product design partners.
Microsoft is introducing a new design for its OneDrive cloud storage aimed at consumers. This upgrade offers visual and functional enhancements to simplify file access and content organization. The revamped interface introduces a people view for finding files through familiar faces, improved filters, and simplified options for file uploads and document creation.