در ادامه تکمیل این ایده از این مقاله و این مقاله هم اینو مطالعه کنید
Background Prompting for Improved Object Depth

https://huggingface.co/papers/2306.05428

#مقاله #ایده_جذاب

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

Transformers as Statisticians

Unveiling a new mechanism "In-Context Algorithm Selection" for In-Context Learning (ICL) in LLMs/transformers.

arxiv.org/abs/2306.04637

#مقاله #ایده_جذاب

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

ده #ایده_جذاب که در هفته گذشته منتشر شد.

1) Tracking Everything Everywhere All at Once - propose a test-time optimization method for estimating dense and long-range motion; enables accurate, full-length motion estimation of every pixel in a video.

2) AlphaDev - a deep reinforcement learning agent which discovers faster sorting algorithms from scratch; the algorithms outperform previously known human benchmarks and have been integrated into the LLVM C++ library.

3) Sparse-Quantized Representation - a new compressed format and quantization technique that enables near-lossless compression of LLMs across model scales; “allows LLM inference at 4.75 bits with a 15% speedup”.

4) MusicGen - a simple and controllable model for music generation built on top of a single-stage transformer LM together with efficient token interleaving patterns; it can be conditioned on textual denoscriptions or melodic features and shows high performance on a standard text-to-music benchmark.

5. Augmenting LLMs with Databases - combines an LLM with a set of SQL databases, enabling a symbolic memory framework; completes tasks via LLM generating SQL instructions that manipulate the DB autonomously.

6) Concept Scrubbing in LLM - presents a method called LEAst-squares Concept Erasure (LEACE) to erase target concept information from every layer in a neural network; it’s used for reducing gender bias in BERT embeddings.

7) Fine-Grained RLHF - trains LMs with fine-grained human feedback; instead of using overall preference, more explicit feedback is provided at the segment level which helps to improve efficacy on long-form question answering, reduce toxicity, and enables LM customization.

8) Hierarchical Vision Transformer - pretrains vision transformers with a visual pretext task (MAE), while removing unnecessary components from a state-of-the-art multi-stage vision transformer; this enables a simple hierarchical vision transformer that’s more accurate and faster at inference and during training.

9) Humor in ChatGPT - explores ChatGPT’s capabilities to grasp and reproduce humor; finds that over 90% of 1008 generated jokes were the same 25 jokes and that ChatGPT is also overfitted to a particular joke structure.

10) Imitating Reasoning Process of Larger LLMs - develops a 13B parameter model that learns to imitate the reasoning process of large foundational models like GPT-4; it leverages large-scale and diverse imitation data and surpasses instruction-tuned models such as Vicuna-13B in zero-shot reasoning.

#مقاله

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

Applications of Transformers

New survey paper highlighting major applications of Transformers for deep learning tasks. Includes a comprehensive list of Transformer models.

arxiv.org/abs/2306.07303

#مقاله

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

Exploring the MIT Mathematics and EECS Curriculum Using LLMs

"GPT-3.5 successfully solves a third of the entire MIT curriculum, while GPT-4, with prompt engineering, achieves a perfect solve rate on a test set excluding questions based on images."

arxiv.org/abs/2306.08997

#مقاله #ایده_جذاب

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

Voicebox: Text-Guided Multilingual Universal Speech Generation at Scale

https://ai.facebook.com/blog/voicebox-generative-ai-model-speech/

Large-scale generative models such as GPT and DALL-E have revolutionized natural language processing and computer vision research. These models not only generate high fidelity text or image outputs, but are also generalists which can solve tasks not explicitly taught. In contrast, speech generative models are still primitive in terms of scale and task generalization. In this paper, we present Voicebox, the most versatile text-guided generative model for speech at scale. Voicebox is a non-autoregressive flow-matching model trained to infill speech, given audio context and text, trained on over 50K hours of speech that are neither filtered nor enhanced. Similar to GPT, Voicebox can perform many different tasks through in-context learning, but is more flexible as it can also condition on future context.

#مقاله #ایده_جذاب

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

Can LLMs Teach Weaker Agents?

Aligned teachers can intervene w/ free-text explanations using Theory of Mind (ExpUtility+Personalization) to improve students on future unexplained data🙂

Misaligned teachers hurt students😢

arxiv.org/abs/2306.09299

#مقاله #ایده_جذاب

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

میخواید اخبار و مقالات و... راجب استارت اپ ها و کمپانیها دریافت کنید اینجا ثبت نام کنید
https://www.joinsuperhuman.ai/subscribe


#خبر

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

رایگان امتحان بدید و رایگان آموزش ببینید

https://lightning.ai/pages/ai-education/deep-learning-fundamentals/

#یادگیری_عمیق #منابع #منابع_پیشنهادی

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

Unifying Large Language Models and Knowledge Graphs: A Roadmap

arxiv.org/abs/2306.08302

#مقاله #ایده_جذاب

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

Sentiment Analysis Of Twitter Data Towards COVID-19 Vaccines Using A Deep Learning Approach

https://ieeexplore.ieee.org/abstract/document/10139297

#مقاله #ایده_جذاب

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

سالانه یک میلیون و هشتصدهزار مقاله منتشر میشه.
محققان هوش مصنوعی برای توضیح و خلاصه کردن مقاله‌ها این راهو معرفی کردند.
https://www.explainpaper.com/

AI explaining AI!

#خبر #هوش_مصنوعی

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

آموزش Data Science دانشگاه هاروارد

1. Lecture notes
2. R code, Python notebooks
3. Lab material
4. Advanced sections

https://harvard-iacs.github.io/2019-CS109A/pages/materials.html

#منابع_پیشنهادی #منابع #علم_داده #آمار

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

Machine Learning with Conformal Prediction for Predictive Maintenance tasks in Industry 4.0

https://www.diva-portal.org/smash/get/diva2:1765779/FULLTEXT01.pdf

https://github.com/valeman/awesome-conformal-prediction

#منابع_پیشنهادی #منابع #یادگیری_ماشین #مقاله

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

چطوری با AI پاورپوینت خودمونو تهیه کنیم؟!

https://twitter.com/itsPaulAi/status/1670061522528137216?t=P04-K8mpYGx0N-kMXYNysg&s=19

#هوش_مصنوعی #خبر

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

در هفته گذشته چه ایده های منتشر شده است:

1) Voicebox - an all-in-one generative speech model; it can synthesize speech across 6 languages; it can perform noise removal, content editing, style conversion, and more; it's 20x faster than current models and outperforms single-purpose models through in-context learning.

2) FinGPT - an open-source LLM for the finance sector; it takes a data-centric approach, providing researchers & practitioners with accessible resources to develop FinLLMs.

3) Crowd Workers Widely Use Large Language Models for Text Production Tasks - estimates that 33-46% of crowd workers on MTurk used LLMs when completing a text production task.

4) Reliability of Watermarks for LLMs - watermarking is useful to detect LLM-generated text and potentially mitigate harms; this work studies the reliability of watermarking for LLMs and finds that watermarks are detectable even when the watermarked text is re-written by humans or paraphrased by another non-watermarked LLM.

5. Applications of Transformers - a new survey paper highlighting major applications of Transformers for deep learning tasks; includes a comprehensive list of Transformer models.

6) Benchmarking NN Training Algorithms - it’s currently challenging to properly assess the best optimizers to train neural networks; this paper presents a new benchmark, AlgoPerf, for benchmarking neural network training algorithms using realistic workloads.

7) Unifying LLMs & Knowledge Graphs - provides a roadmap for the unification of LLMs and KGs; covers how to incorporate KGs in LLM pre-training/inferencing, leverage LLMs for KG tasks such as question answering, and enhance both KGs and LLMs for bidirectional reasoning.

8) Augmenting LLMs with Long-term Memory - proposes a framework to enable LLMs to memorize long history; it’s enhanced with memory-augmented adaptation training to memorize long past context and use long-term memory for language modeling; achieves improvements on memory-augmented in-context learning over LLMs.

9) TAPIR - enables tracking any queried point on any physical surface throughout a video sequence; outperforms all baselines and facilitates fast inference on long and high-resolution videos (track points faster than real-time when using modern GPUs).

10) Mind2Web - a new dataset for evaluating generalist agents for the web; contains 2350 tasks from 137 websites over 31 domains; it enables testing generalization ability across tasks and environments, covering practical use cases on the web.

#مقاله #ایده_جذاب

🔸 مطالب بیشتر 👇👇

✅ @AI_DeepMind

Demystifying GPT Self-Repair for Code Generation

We've seen a couple of papers showing the promise of self-repair in code generation. This paper finds that modest performance gains are seen when using GPT-4 for textual feedback.

Another interesting finding: significant performance is seen when GPT-4 provides feedback to GPT-3.5 and expert human programmers provide feedback to programs generated by GPT-4.

Feedback seems to be crucial for self-repair but I sense there is a lot more work to be done when using model-generated feedback. Human feedback is still a strong approach and more investigation is needed to figure out when to rely on human intervention.

arxiv.org/abs/2306.09896