Zobrazeno 1 - 10
of 79
pro vyhledávání: '"Bhardwaj, Rishabh"'
Autor:
Gupta, Prannaya, Yau, Le Qi, Low, Hao Han, Lee, I-Shiang, Lim, Hugo Maximus, Teoh, Yu Xin, Koh, Jia Hng, Liew, Dar Win, Bhardwaj, Rishabh, Bhardwaj, Rajat, Poria, Soujanya
WalledEval is a comprehensive AI safety testing toolkit designed to evaluate large language models (LLMs). It accommodates a diverse range of models, including both open-weight and API-based ones, and features over 35 safety benchmarks covering areas
Externí odkaz:
http://arxiv.org/abs/2408.03837
We propose Ruby Teaming, a method that improves on Rainbow Teaming by including a memory cache as its third dimension. The memory dimension provides cues to the mutator to yield better-quality prompts, both in terms of attack success rate (ASR) and q
Externí odkaz:
http://arxiv.org/abs/2406.11654
With the proliferation of domain-specific models, model merging has emerged as a set of techniques that combine the capabilities of multiple models into one that can multitask without the cost of additional training. In this paper, we propose a new m
Externí odkaz:
http://arxiv.org/abs/2406.11617
We use partial wave unitarity to constrain various bespoke four-point amplitudes. We start by constructing bespoke generalizations of the type I superstring amplitude, which we show satisfy dual resonance and have suitable high-energy limits. By anal
Externí odkaz:
http://arxiv.org/abs/2406.04410
Neural speech synthesis, or text-to-speech (TTS), aims to transform a signal from the text domain to the speech domain. While developing TTS architectures that train and test on the same set of speakers has seen significant improvements, out-of-domai
Externí odkaz:
http://arxiv.org/abs/2404.04645
Conformally soft operators and their associated soft theorems on the celestial sphere encode the low energy behaviour of bulk scattering amplitudes. They lead to an infinite dimensional symmetry algebra of the celestial CFT at tree-level. In this pap
Externí odkaz:
http://arxiv.org/abs/2403.10443
Aligned language models face a significant limitation as their fine-tuning often results in compromised safety. To tackle this, we propose a simple method RESTA that performs LLM safety realignment. RESTA stands for REstoring Safety through Task Arit
Externí odkaz:
http://arxiv.org/abs/2402.11746
We study the twisted (co)homology of a family of genus-one integrals -- the so called Riemann-Wirtinger integrals. These integrals are closely related to one-loop string amplitudes in chiral splitting where one leaves the loop-momentum, modulus and a
Externí odkaz:
http://arxiv.org/abs/2312.02148
Adapters are widely popular parameter-efficient transfer learning approaches in natural language processing that insert trainable modules in between layers of a pre-trained language model. Apart from several heuristics, however, there has been a lack
Externí odkaz:
http://arxiv.org/abs/2310.19232
Autor:
Bhardwaj, Rishabh, Poria, Soujanya
Red-teaming has been a widely adopted way to evaluate the harmfulness of Large Language Models (LLMs). It aims to jailbreak a model's safety behavior to make it act as a helpful agent disregarding the harmfulness of the query. Existing methods are pr
Externí odkaz:
http://arxiv.org/abs/2310.14303