-
2024Modern language models (LMs) need to follow human instructions while being faithful; yet, they often fail to achieve both. Here, we provide concrete evidence of a trade-off between instruction following (i.e., follow open-ended instructions) and faithfulness (i.e., ground responses in given context) when training LMs with these objectives. For instance, fine-tuning LLaMA-7B on instruction following datasets
-
2024Question answering based on retrieval-augmented generation (RAG-QA) is an important research topic in NLP and has a wide range of real-world applications. However, most existing datasets for this task are either constructed using a single source corpus or consist of short extractive answers, which fall short of evaluating large language model (LLM) based RAG-QA systems on cross-domain generalization. To
-
2024Learning of preference models from human feedback has been central to recent advances in artificial intelligence. Motivated by the cost of obtaining high-quality human annotations, we study efficient human preference elicitation for learning preference models. The key idea in our work is to generalize optimal designs, a methodology for computing optimal information-gathering policies, to questions with
-
2024Large language model advancements have enabled the development of multi-agent frameworks to tackle complex, real-world problems such as to automate tasks that require interactions with diverse tools, reasoning, and human collaboration. We present MARCO, a Multi-Agent Real-time Chat Orchestration framework for automating tasks using LLMs. MARCO addresses key challenges in utilizing LLMs for complex, multi-step
-
2024Training large AI models such as deep learning recommendation systems and large language models (LLMs) costs massive GPUs and computing time. The high training cost has become only affordable to big tech companies, meanwhile also causing increasing concerns about the environmental impact. This paper presents CoMERA, a Computing- and Memory-Efficient training method via Rank-Adaptive tensor optimization.
Related content
-
July 09, 2023Finding that 70% of attention heads and 20% of feed-forward networks can be excised with minimal effect on in-context learning suggests that large language models are undertrained.
-
July 07, 2023Amazon’s Yang Liu, general chair of this year’s meeting of the Association for Computational Linguistics, on the road ahead for LLMs.
-
July 06, 2023The program exposes students to computer science as they create their own Alexa skills.
-
July 05, 2023Amazon Research Award recipient Shrikanth Narayanan is on a mission to make inclusive human-AI conversational experiences.
-
July 03, 2023With little training data and no mapping of speech to phonemes, Amazon researchers used voice conversion to generate Irish-accented training data in Alexa’s own voice.
-
June 26, 2023How phonetically blended results (PBR) help ensure customers find the content they were actually asking for.