🔥 Trending AI Research: Let’s learn something new from the trending papers.

💻 Some Cool Github Repos: Take a deep dive into the world of advanced AI with these trending Github repos

🏃 Let’s Practice Some Generative AI: Let’s do some fun in the lab by trying some prompts and AI tools.

🛎️ Trending Tools: Check out some cool AI tools picked up by our editorial team.

Read Time: 5 Minutes

🔥Trending AI Research

1️⃣ Meet Magic123, a two-stage coarse-to-fine approach for high-quality, textured 3D meshes generation from a single unposed image in the wild using both 2D and 3D priors:

By leveraging both 2D and 3D priors, this approach overcomes the limitations of existing studies and achieves state-of-the-art results in image-to-3D reconstruction. The trade-off parameter between the 2D and 3D priors allows for control over the balance between exploration and exploitation of the generated geometry. This method outperforms previous techniques in terms of both realism and level of detail, as demonstrated through extensive experiments on real-world images and synthetic benchmarks. The findings contribute to narrowing the gap between human abilities in 3D reasoning and those of machines and pave the way for future advancements in single-image 3D reconstruction.

2️⃣ This AI paper from Alibaba proposes Preference Ranking Optimization (PRO) as an alternative to PPO for directly aligning LLMs with the Bradley-Terry comparison.

PRO extends the pairwise Bradley-Terry comparison to accommodate preference rankings of any length. By iteratively contrasting the likelihood of generating responses, PRO instructs the LLM to prioritize the best response while progressively ranking the remaining responses. In this manner, PRO effectively transforms human alignment into aligning the probability ranking of n responses generated by LLM with the preference ranking of humans towards these responses. According to the authors of the paper, experiments have shown that PRO outperforms existing alignment algorithms, achieving comparable results to ChatGPT and human responses through automatic-based, reward-based, GPT-4, and human evaluations.

3️⃣ Google Researchers propose to significantly reduce the burden on LLMs by using a new technique called Pairwise Ranking Prompting (PRP).

According to the authors, the published results are the first in the literature to achieve state-of-the-art ranking performance on standard benchmarks using moderate-sized open-sourced LLMs. On TREC-DL2020, PRP based on the Flan-UL2 model with 20B parameters outperforms the previous best approach in the literature, which is based on the blackbox commercial GPT-4 that has 50x (estimated) model size, by over 5% at NDCG@1. On TREC-DL2019, PRP is only inferior to the GPT-4 solution on the NDCG@5 and NDCG@10 metrics, while outperforming other existing solutions, such as InstructGPT which has 175B parameters, by over 10% for nearly all ranking metrics.

4️⃣ Microsoft researchers introduce a Pareto optimal self-supervision framework that can leverage available programmatic supervision to systematically calibrate LLM responses by producing a risk score for every response, without any additional manual efforts.

This is accomplished by learning a harmonizer model to align LLM output with other available supervision sources, which would assign higher risk scores to more uncertain LLM responses and facilitate error correction. They also proposed a Pareto optimal learning assessed risk (POLAR) score which gets consistent and well-calibrated to the probability of LLM error.

💻 Github Repos

➡️ CASIA-IVA-Lab / FastSAM: The Fast Segment Anything Model (FastSAM) is a CNN Segment Anything Model trained by only 2% of the SA-1B dataset published by SAM authors. The FastSAM achieve a comparable performance with the SAM method at 50× higher run-time speed.

➡️ chat2db / Chat2DB: An intelligent and versatile general-purpose SQL client and reporting tool for databases that integrates ChatGPT capabilities.

➡️ ramonvc / freegpt-webui: This project features a WebUI utilizing the G4F API. Experience the power of ChatGPT with a user-friendly interface, enhanced jailbreaks, and completely free.

➡️ vllm-project / vllm: vLLM is a fast and easy-to-use library for LLM inference and serving.

➡️ turboderp / exllama: A standalone Python/C++/CUDA implementation of Llama for use with 4-bit GPTQ weights, designed to be fast and memory-efficient on modern GPUs.

🏃Let’s Practice Some Generative AI

With This AI Tool, You Can Ask Questions To Your CSV File. Introducing ChatCSV, your newest personal data analyst primed to revolutionize your data processing.

Let's do some exercises with ChatCSV

Step 1: Open https://www.chatcsv.co/

Step 2: Login or Signup

Step 3: Now lets download a CSV File from Kaggle for 'World Happiness Report'

Step 4: Take any of the CSV files from the download zip folder. I am taking 2018 CSV File.

Step 5: Upload this file on https://www.chatcsv.co/ after you logged in

Step 6: Start Asking Questions in the chat

🛎️ Trending Tools

Bardeen: Automate your manual and repetitive tasks using AI, and focus on the ones that actually matter.

AdCreative AI: Generate conversion-focused ad creatives and social media post creatives in a matter of seconds using Artificial Intelligence

Booth AI: Experience the extraordinary with Booth.ai: AI-generated photos that redefine product photography. Effortlessly create pro-quality lifestyle shots by specifying your vision and uploading sample product images.

Sheet+: Unleash Sheet+, the formula genius! This AI-powered tool speeds up Google Sheets and Excel, transforming text into precise formulas in a flash.

Slayer AI: Slayer AI revolutionizes podcasting by crafting tailor-made shows in mere seconds.

Sponsored Section

🤝 Partner with us