Dissociating language and thought in large language models
Large language models (LLMs) have come closest among all models to date to mastering
human language, yet opinions about their linguistic and cognitive capabilities remain split …
human language, yet opinions about their linguistic and cognitive capabilities remain split …
Challenges and applications of large language models
Large Language Models (LLMs) went from non-existent to ubiquitous in the machine
learning discourse within a few years. Due to the fast pace of the field, it is difficult to identify …
learning discourse within a few years. Due to the fast pace of the field, it is difficult to identify …
A survey of large language models
Language is essentially a complex, intricate system of human expressions governed by
grammatical rules. It poses a significant challenge to develop capable AI algorithms for …
grammatical rules. It poses a significant challenge to develop capable AI algorithms for …
Siren's song in the AI ocean: a survey on hallucination in large language models
While large language models (LLMs) have demonstrated remarkable capabilities across a
range of downstream tasks, a significant concern revolves around their propensity to exhibit …
range of downstream tasks, a significant concern revolves around their propensity to exhibit …
Inference-time intervention: Eliciting truthful answers from a language model
Abstract We introduce Inference-Time Intervention (ITI), a technique designed to enhance
the" truthfulness" of large language models (LLMs). ITI operates by shifting model activations …
the" truthfulness" of large language models (LLMs). ITI operates by shifting model activations …
Rwkv: Reinventing rnns for the transformer era
Transformers have revolutionized almost all natural language processing (NLP) tasks but
suffer from memory and computational complexity that scales quadratically with sequence …
suffer from memory and computational complexity that scales quadratically with sequence …
Erasing concepts from diffusion models
R Gandikota, J Materzynska… - Proceedings of the …, 2023 - openaccess.thecvf.com
Motivated by concerns that large-scale diffusion models can produce undesirable output
such as sexually explicit content or copyrighted artistic styles, we study erasure of specific …
such as sexually explicit content or copyrighted artistic styles, we study erasure of specific …
Talking about large language models
M Shanahan - Communications of the ACM, 2024 - dl.acm.org
Talking about Large Language Models Page 1 key insights ˽ As LLMs become more powerful,
it becomes increasingly tempting to describe LLM-based dialog agents in human-like terms …
it becomes increasingly tempting to describe LLM-based dialog agents in human-like terms …
Towards automated circuit discovery for mechanistic interpretability
Through considerable effort and intuition, several recent works have reverse-engineered
nontrivial behaviors oftransformer models. This paper systematizes the mechanistic …
nontrivial behaviors oftransformer models. This paper systematizes the mechanistic …
Mass-editing memory in a transformer
Recent work has shown exciting promise in updating large language models with new
memories, so as to replace obsolete information or add specialized knowledge. However …
memories, so as to replace obsolete information or add specialized knowledge. However …