paper-with-me

홈 › Papers

Capturing AI's Attention: Physics of Repetition, Hallucination, Bias and Beyond

2025-04-06 · Frank Yingjie Huo, Neil F. Johnson

We derive a first-principles physics theory of the AI engine at the heart of LLMs' 'magic' (e.g. ChatGPT, Claude): the basic Attention head. The theory allows a quantitative analysis of outstanding AI challenges such as output repetition, hallucination and harmful content, and bias (e.g. from training and fine-tuning). Its predictions are consistent with large-scale LLM outputs. Its 2-body form suggests why LLMs work so well, but hints that a generalized 3-body Attention would make such AI work even better. Its similarity to a spin-bath means that existing Physics expertise could immediately be harnessed to help Society ensure AI is trustworthy and resilient to manipulation.

📄 PDF Abstract BibTeX arXiv:2504.04600

Code (0)

등록된 구현이 없습니다.

Tasks

Hallucination

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

Testing the spin-bath view of self-attention: A Hamiltonian analysis of GPT-2 Transformer

2025-07-01 · Satadeep Bhattacharjee, Seung-Cheol Lee arxiv

The recently proposed physics-based framework by Huo and Johnson~\cite{huo2024capturing} models the attention mechanism of Large Language Models (LLMs) as an interacting two-body spin system, offering a first-principles …

Beyond Transcription: Mechanistic Interpretability in ASR

2025-08-21 · Neta Glazer, Yael Segal-Feldman, Hilit Segev, Aviv Shamsian 외 arxiv

Interpretability methods have recently gained significant attention, particularly in the context of large language models, enabling insights into linguistic representations, error detection, and model behaviors such as h…

Speech Recognition

Mitigating Hallucinations in Zero-Shot Scientific Summarisation: A Pilot Study

2025-11-30 · Imane Jaaouine, Ross D. King arxiv

Large language models (LLMs) produce context inconsistency hallucinations, which are LLM generated outputs that are misaligned with the user prompt. This research project investigates whether prompt engineering (PE) meth…

Prompt Engineering

From Loops to Oops: Fallback Behaviors of Language Models Under Uncertainty

2024-07-08 · Maor Ivgi, Ori Yoran, Jonathan Berant, Mor Geva

Large language models (LLMs) often exhibit undesirable behaviors, such as hallucinations and sequence repetitions. We propose to view these behaviors as fallbacks that models exhibit under uncertainty, and investigate th…

Instruction Following

Do LLM Decoders Listen Fairly? Benchmarking How Language Model Priors Shape Bias in Speech Recognition

2026-04-23 · Srishti Ginjala, Eric Fosler-Lussier, Christopher W. Myers, Srinivasan Parthasarathy arxiv

As pretrained large language models replace task-specific decoders in speech recognition, a critical question arises: do their text-derived priors make recognition fairer or more biased across demographic groups? We eval…

Speech Recognition