Zobrazeno 1 - 10
of 38
pro vyhledávání: '"Josifoski A"'
Recent work demonstrated great promise in the idea of orchestrating collaborations between LLMs, human input, and various tools to address the inherent limitations of LLMs. We propose a novel perspective called semantic decoding, which frames these c
Externí odkaz:
http://arxiv.org/abs/2403.14562
Autor:
Amani, Mohammad Hossein, Baldwin, Nicolas Mario, Mansouri, Amin, Josifoski, Martin, Peyrard, Maxime, West, Robert
Traditional language models, adept at next-token prediction in text sequences, often struggle with transduction tasks between distinct symbolic systems, particularly when parallel data is scarce. Addressing this issue, we introduce \textit{symbolic a
Externí odkaz:
http://arxiv.org/abs/2402.10575
Constrained decoding, a technique for enforcing constraints on language model outputs, offers a way to control text generation without retraining or architectural modifications. Its application is, however, typically restricted to models that give us
Externí odkaz:
http://arxiv.org/abs/2401.09967
Autor:
Davidson, Tim R., Veselovsky, Veniamin, Josifoski, Martin, Peyrard, Maxime, Bosselut, Antoine, Kosinski, Michal, West, Robert
We introduce an approach to evaluate language model (LM) agency using negotiation games. This approach better reflects real-world use cases and addresses some of the shortcomings of alternative LM benchmarks. Negotiation games enable us to study mult
Externí odkaz:
http://arxiv.org/abs/2401.04536
Autor:
Monea, Giovanni, Peyrard, Maxime, Josifoski, Martin, Chaudhary, Vishrav, Eisner, Jason, Kıcıman, Emre, Palangi, Hamid, Patra, Barun, West, Robert
Large language models (LLMs) have an impressive ability to draw on novel information supplied in their context. Yet the mechanisms underlying this contextual grounding remain unknown, especially in situations where contextual information contradicts
Externí odkaz:
http://arxiv.org/abs/2312.02073
Autor:
Josifoski, Martin, Klein, Lars, Peyrard, Maxime, Baldwin, Nicolas, Li, Yifei, Geng, Saibo, Schnitzler, Julian Paul, Yao, Yuxing, Wei, Jiheng, Paul, Debjit, West, Robert
Recent advances in artificial intelligence (AI) have produced highly capable and controllable systems. This creates unprecedented opportunities for structured reasoning as well as collaboration among multiple AI systems and humans. To fully realize t
Externí odkaz:
http://arxiv.org/abs/2308.01285
Autor:
Veselovsky, Veniamin, Ribeiro, Manoel Horta, Arora, Akhil, Josifoski, Martin, Anderson, Ashton, West, Robert
Large Language Models (LLMs) have democratized synthetic data generation, which in turn has the potential to simplify and broaden a wide gamut of NLP tasks. Here, we tackle a pervasive problem in synthetic data generation: its generative distribution
Externí odkaz:
http://arxiv.org/abs/2305.15041
Despite their impressive performance, large language models (LMs) still struggle with reliably generating complex output structures when not finetuned to follow the required output format exactly. To address this issue, grammar-constrained decoding (
Externí odkaz:
http://arxiv.org/abs/2305.13971
Large language models (LLMs) have great potential for synthetic data generation. This work shows that useful data can be synthetically generated even for tasks that cannot be solved directly by LLMs: for problems with structured outputs, it is possib
Externí odkaz:
http://arxiv.org/abs/2303.04132
Publikováno v:
Journal of Machine Learning Research (24), 2023, 1-62
Meta-Learning aims to speed up the learning process on new tasks by acquiring useful inductive biases from datasets of related learning tasks. While, in practice, the number of related tasks available is often small, most of the existing approaches a
Externí odkaz:
http://arxiv.org/abs/2211.07206