• AI Research Insights
  • Posts
  • ↗️ AI/ML Research Updates: Ghostbuster (A SOTA AI Method for Detecting LLM-Generated Text); Tencent AI Lab Introduces Chain-of-Noting (CoN); Stanford University Researchers Introduce FlashFFTConv; and many more research trends

↗️ AI/ML Research Updates: Ghostbuster (A SOTA AI Method for Detecting LLM-Generated Text); Tencent AI Lab Introduces Chain-of-Noting (CoN); Stanford University Researchers Introduce FlashFFTConv; and many more research trends

This newsletter brings AI research news that is much more technical than most resources but still digestible and applicable

Hey Folks!

This newsletter will discuss some cool AI research papers and AI tools. Happy learning!

👉 What is Trending in AI/ML Research?

How can we effectively detect AI-generated text? This paper introduces "Ghostbuster," a novel system designed for this purpose. Ghostbuster operates by channeling documents through several weaker language models, conducting a structured search across the feature combinations they produce, and then training a classifier on these features to discern if the text is AI-generated. A key advantage of Ghostbuster is its independence from the token probabilities of the target model, allowing it to identify texts from black-box models or unknown versions. Accompanying the model are three new datasets in the realms of student essays, creative writing, and news articles, serving as benchmarks for detection. Ghostbuster's performance surpasses existing detectors like DetectGPT and GPTZero, including a new RoBERTa baseline, achieving a 99.0 F1 score across various domains. This score is notably 5.9 F1 points higher than the best existing models. The system demonstrates superior generalization across writing domains, prompting strategies, and language models. It also exhibits robustness against various perturbations and paraphrasing attacks and is effective with texts from non-native English speakers.

How can Retrieval-Augmented Language Models (RALMs) be enhanced to address the issue of unreliable information retrieval? This paper introduces Chain-of-Noting (CoN), an innovative method designed to bolster RALMs against irrelevant or noisy data, and to improve their handling of unknown scenarios. CoN operates by generating sequential reading notes for each retrieved document, assessing their relevance to the query, and integrating this assessment into the response process. Trained using data created by ChatGPT and then implemented on an LLaMa-2 7B model, CoN demonstrates a significant performance improvement over standard RALMs. In tests across four open-domain QA benchmarks, CoN showed an average increase of +7.9 in EM score with completely noisy documents and a +10.5 increase in rejection rates for questions beyond the model's pre-training knowledge.

How can convolution models with long filters, crucial in long-sequence tasks, compete with the speed of optimized Transformers? This paper introduces "FlashFFTConv," a solution optimizing FFT (Fast Fourier Transform) convolutions to address their poor hardware utilization and high memory hierarchy I/O. FlashFFTConv employs matrix decomposition for FFT execution using matrix multiply units and kernel fusion for long sequences, thus reducing I/O. It also introduces two sparse convolution algorithms: partial and frequency-sparse convolutions, achieved by skipping blocks in matrix decomposition, further saving memory and compute resources. The results are impressive: FlashFFTConv accelerates FFT convolutions by up to 7.93× over PyTorch and achieves up to 4.4× end-to-end speedup. This enhancement allows models like Hyena-GPT-s and M2-BERT-base to outperform larger counterparts in perplexity and GLUE score, respectively. Remarkably, FlashFFTConv also achieves unprecedented accuracy in high-resolution vision tasks and enables the first DNA model to process the longest human genes, showcasing its potential in a variety of applications.

How can in-document search be enhanced with external knowledge sources for more efficient information access? This paper introduces the "KTRL+F" task, addressing the challenge of real-time semantic target identification within a document, augmented by external knowledge. Traditional methods struggle with hallucinations, latency, and effectively leveraging external knowledge. To overcome these, the authors propose a Knowledge-Augmented Phrase Retrieval model, which integrates external knowledge embeddings into phrase embeddings, striking a balance between speed and accuracy. A user study confirms the model's effectiveness, showing users could search faster with fewer queries and less reliance on external information sources. This paper invites further research in KTRL+F to advance in-document search efficiency.

Featured AI Tools For You

  • SaneBox*: SaneBox: AI-powered email management that saves you time and brings sanity back to your inbox. Voted Best Productivity Apps for 2023 on PCMag. Sign up today and save $25 on any subscription. [Email and Productivity]

  • Adcreative AI*: Boost your advertising and social media game with AdCreative.ai - the ultimate Artificial Intelligence solution. [Marketing and Sales]

  • Mubert*: Mubert is an AI-powered platform that lets you create personalized soundtracks and tunes. [Music]

  • Retouch4me: Retouch4me's plugins make photo retouching such a breeze, ensuring professional results every time. [Photo Editing]

  • Parsio*: Automate your data extraction using our AI-powered PDF parser and eliminate manual data entry. [OCR-PDF and Productivity]

  • Tugan.ai*: Tugan AI turns articles, videos, and sales pages into engaging content for newsletters, social media, and email. [Social Media and Marketing]

*We do make a small affiliate profit when you buy this product through the click link

Sponsorship: For newsletter sponsorship, please reach us at [email protected]

Marktechpost Media Inc. is a California-based Artificial Intelligence News Platform with 2 Million+ AI Tech Readers/Viewers

Who is Marktechpost's Audience?

Our audience consists of Data Engineers, MLOps Engineers, Data Scientists, ML Engineers, ML Researchers, Data Analysts, Software Developers, Architects, IT Managers, Software engineer/SDEs, CTO, Director/ VP data science, CEOs, PhD Researchers, Postdocs and Tech Investors.

Who should try our Advertisement or Sponsorship package?

We encourage companies who are developing AI software, Deep learning tools, Machine learning tools, NLP tools, MLOps tools, Data Science tools, AIOps, DataOps, BigData tools, AI Chips/Hardware, GPUs, TPUs, CPUs, and SaaS products.