Tech Frontier

Apr 05 2024 33 ep. 1

Tech Frontier Podcast artwork

"Tech Frontier" is your daily digest for AI research. Tailored for AI researchers and engineers, this podcast delivers succinct summaries of cutting-edge papers, keeping you informed and ahead in the fast-paced AI landscape. Tune in daily to stay on top of the latest in AI, effortlessly.

Technology Science

Copy RSS

Subscribe on Podcast Addict

CoMat: Aligning Text-to-Image Diffusion Model with Image-to-Text Concept Matching

Apr 05 2024 2 mins

Diffusion models have demonstrated great success in the field of text-to-image generation. However, alleviating the misalignment between the text prompts and images is still challenging. The root reason behind the misalignment has not been extensively investigated. We observe that the misalignment is caused by inadequate token attention activation. We further attribute this phenome

AutoWebGLM: Bootstrap And Reinforce A Large Language Model-based Web Navigating Agent

Apr 05 2024 3 mins

Large language models (LLMs) have fueled many intelligent agent tasks, such as web navigation -- but most existing agents perform far from satisfying in real-world webpages due to three factors: (1) the versatility of actions on webpages, (2) HTML text exceeding model processing capacity, and (3) the complexity of decision-making due to the open-domain nature of web. In light of th

Training LLMs over Neurally Compressed Text

Apr 05 2024 2 mins

In this paper, we explore the idea of training large language models (LLMs) over highly compressed text. While standard subword tokenizers compress text by a small factor, neural text compressors can achieve much higher rates of compression. If it were possible to train LLMs directly over neurally compressed text, this would confer advantages in training and serving efficiency, as

ReFT: Representation Finetuning for Language Models

Apr 05 2024 2 mins

Parameter-efficient fine-tuning (PEFT) methods seek to adapt large models via updates to a small number of weights. However, much prior interpretability work has shown that representations encode rich semantic information, suggesting that editing representations might be a more powerful alternative. Here, we pursue this hypothesis by developing a family of Representation Finetuning

PointInfinity: Resolution-Invariant Point Diffusion Models

Apr 05 2024 2 mins

We present PointInfinity, an efficient family of point cloud diffusion models. Our core idea is to use a transformer-based architecture with a fixed-size, resolution-invariant latent representation. This enables efficient training with low-resolution point clouds, while allowing high-resolution point clouds to be generated during inference. More importantly, we show that scaling th

Red Teaming GPT-4V: Are GPT-4V Safe Against Uni/Multi-Modal Jailbreak Attacks?

Apr 05 2024 3 mins

Various jailbreak attacks have been proposed to red-team Large Language Models (LLMs) and revealed the vulnerable safeguards of LLMs. Besides, some methods are not limited to the textual modality and extend the jailbreak attack to Multimodal Large Language Models (MLLMs) by perturbing the visual input. However, the absence of a universal evaluation benchmark complicates the perform

LVLM-Intrepret: An Interpretability Tool for Large Vision-Language Models

Apr 05 2024 3 mins

In the rapidly evolving landscape of artificial intelli-gence, multi-modal large language models are emerging asa significant area of interest. These models, which combinevarious forms of data input, are becoming increasingly pop-ular. However, understanding their internal mechanisms re-mains a complex task. Numerous advancements have beenmade in the field of explainability tools a

Mixture-of-Depths: Dynamically allocating compute in transformer-based language models

Apr 04 2024 2 mins

Transformer-based language models spread FLOPs uniformly across input sequences. In this work we demonstrate that transformers can instead learn to dynamically allocate FLOPs (or compute) to specific positions in a sequence, optimising the allocation along the sequence for different layers across the model depth. Our method enforces a total compute budget by capping the number of t

Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models

Apr 04 2024 3 mins

Algorithmic reasoning refers to the ability to understand the complex patterns behind the problem and decompose them into a sequence of reasoning steps towards the solution. Such nature of algorithmic reasoning makes it a challenge for large language models (LLMs), even though they have demonstrated promising performance in other reasoning tasks. Within this context, some recent st

Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction

Apr 04 2024 4 mins

We present Visual AutoRegressive modeling (VAR), a new generation paradigm that redefines the autoregressive learning on images as coarse-to-fine "next-scale prediction" or "next-resolution prediction", diverging from the standard raster-scan "next-token prediction". This simple, intuitive methodology allows autoregressive (AR) transformers to learn visual distributions fast and ge