@_akhaliq - AI Dynamics - Page 128 of 138

Generating Diverse Text Data with LLMs and Human Input

By

–

08 June 2023 3h29

Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventions paper page: https://
huggingface.co/papers/2306.04
140
… Large language models (LLMs) can be used to generate text data for training and evaluating other models. However,

→ View original post on X — @_akhaliq

8 June 2023

Text-only Domain Adaptation for Speech Recognition using Unified Representation

By

@_akhaliq

–

08 June 2023 3h26

Text-only Domain Adaptation using Unified Speech-Text Representation in Transducer paper page: https://
huggingface.co/papers/2306.04
076
… Domain adaptation using text-only corpus is challenging in end-to-end(E2E) speech recognition. Adaptation by synthesizing audio from text through TTS is

→ View original post on X — @_akhaliq

8 June 2023

Certified Reasoning with Language Models for Sound Step-by-Step Reasoning

By

@_akhaliq

–

08 June 2023 3h24

Certified Reasoning with Language Models paper page: https://
huggingface.co/papers/2306.04
031
… Language models often achieve higher accuracy when reasoning step-by-step in complex tasks. However, their reasoning can be unsound, inconsistent, or rely on undesirable prior assumptions. To tackle

→ View original post on X — @_akhaliq

8 June 2023

Multi-Hop Reasoning in Language Models via Soft Prompts

By

@_akhaliq

–

08 June 2023 3h11

Triggering Multi-Hop Reasoning for Question Answering in Language Models using Soft Prompts and Random Walks paper page: https://
huggingface.co/papers/2306.04
009
… Despite readily memorizing world knowledge about entities, pre-trained language models (LMs) struggle to compose together two or more

→ View original post on X — @_akhaliq

8 June 2023

M3IT: Large-Scale Multi-Modal Multilingual Instruction Tuning Dataset

By

@_akhaliq

–

08 June 2023 2h47

M3IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning paper page: https://
huggingface.co/papers/2306.04
387
… Instruction tuning has significantly advanced large language models (LLMs) such as ChatGPT, enabling them to align with human instructions across diverse tasks.

→ View original post on X — @_akhaliq

8 June 2023

Vision Pro and AI/VR Tools: The Next Big Opportunities

By

@_akhaliq

–

07 June 2023 18h45

vision pro is the next bug thing ChatGPT is dead I found 1 million AI/VR tools released this week to help you make a bajillion dollars Don’t get left behind

→ View original post on X — @_akhaliq

7 June 2023

InternLM: Multilingual 104B Parameter Language Model Released

By

@_akhaliq

–

07 June 2023 15h46

InternLM: A Multilingual Language Model with Progressively Enhanced Capabilities github: https://
github.com/InternLM/Inter
nLM-techreport
… present InternLM, a multilingual foundational language model with 104B parameters. InternLM is pre-trained on a large corpora with 1.6T tokens with a multi-phase

→ View original post on X — @_akhaliq

7 June 2023

Recognize Anything: Strong Foundation Model for Image Tagging

By

@_akhaliq

–

07 June 2023 4h38

Recognize Anything: A Strong Image Tagging Model paper page: https://
huggingface.co/papers/2306.03
514
…
demo: https://
huggingface.co/spaces/xinyu12
05/Tag2Text
… present the Recognize Anything Model (RAM): a strong foundation model for image tagging. RAM can recognize any common category with high accuracy. RAM introduces a

→ View original post on X — @_akhaliq

7 June 2023