The rapid advancement and deployment of AI systems have created an urgent need for standard safety-evaluation frameworks. This paper introduces AILuminate v1.0, the first comprehensive industry-standard benchmark for assessing AI-product risk and reliability. Its development employed an open process that included participants from multiple fields. The benchmark evaluates an AI system's resistance to prompts designed to elicit dangerous, illegal, or undesirable behavior in 12 hazard categories, including violent crimes, nonviolent crimes, sex-related crimes, child sexual exploitation, indiscriminate weapons, suicide and self-harm, intellectual property, privacy, defamation, hate, sexual content, and specialized advice (election, financial, health, legal). Our method incorporates a complete assessment standard, extensive prompt datasets, a novel evaluation framework, a grading and reporting system, and the technical as well as organizational infrastructure for long-term support and evolution. In particular, the benchmark employs an understandable five-tier grading scale (Poor to Excellent) and incorporates an innovative entropy-based system-response evaluation. In addition to unveiling the benchmark, this report also identifies limitations of our method and of building safety benchmarks generally, including evaluator uncertainty and the constraints of single-turn interactions. This work represents a crucial step toward establishing global standards for AI risk and reliability evaluation while acknowledging the need for continued development in areas such as multiturn interactions, multimodal understanding, coverage of additional languages, and emerging hazard categories. Our findings provide valuable insights for model developers, system integrators, and policymakers working to promote safer AI deployment.

Fine-tuning large language models (LLMs) for downstream tasks is a widely adopted approach, but it often leads to safety degradation in safety-aligned LLMs. Currently, many solutions address this issue by incorporating additional safety data, which can be impractical in many cases. In this paper, we address the question: How can we improve downstream task performance while preserving safety in LLMs without relying on additional safety data? We propose a simple and effective method that maintains the inherent safety of LLMs while enhancing their downstream task performance: merging the weights of pre- and post-fine-tuned safety-aligned models. Experimental results across various downstream tasks, models, and merging methods demonstrate that this approach effectively mitigates safety degradation while improving downstream task performance, offering a practical solution for adapting safety-aligned LLMs.

In this work we explore utilizing LLMs for data augmentation for manufacturing task guidance system. The dataset consists of representative samples of interactions with technicians working in an advanced manufacturing setting. The purpose of this work to explore the task, data augmentation for the supported tasks and evaluating the performance of the existing LLMs. We observe that that task is complex requiring understanding from procedure specification documents, actions and objects sequenced temporally. The dataset consists of 200,000+ question/answer pairs that refer to the spec document and are grounded in narrations and/or video demonstrations. We compared the performance of several popular open-sourced LLMs by developing a "baseline" using each LLM and then compared the responses in a reference-free setting using LLM-as-a-judge and compared the ratings with crowd-workers whilst validating the ratings with experts.

This paper introduces v0.5 of the AI Safety Benchmark, which has been created by the MLCommons AI Safety Working Group. The AI Safety Benchmark has been designed to assess the safety risks of AI systems that use chat-tuned language models. We introduce a principled approach to specifying and constructing the benchmark, which for v0.5 covers only a single use case (an adult chatting to a general-purpose assistant in English), and a limited set of personas (i.e., typical users, malicious users, and vulnerable users). We created a new taxonomy of 13 hazard categories, of which 7 have tests in the v0.5 benchmark. We plan to release version 1.0 of the AI Safety Benchmark by the end of 2024. The v1.0 benchmark will provide meaningful insights into the safety of AI systems. However, the v0.5 benchmark should not be used to assess the safety of AI systems. We have sought to fully document the limitations, flaws, and challenges of v0.5. This release of v0.5 of the AI Safety Benchmark includes (1) a principled approach to specifying and constructing the benchmark, which comprises use cases, types of systems under test (SUTs), language and context, personas, tests, and test items; (2) a taxonomy of 13 hazard categories with definitions and subcategories; (3) tests for seven of the hazard categories, each comprising a unique set of test items, i.e., prompts. There are 43,090 test items in total, which we created with templates; (4) a grading system for AI systems against the benchmark; (5) an openly available platform, and downloadable tool, called ModelBench that can be used to evaluate the safety of AI systems on the benchmark; (6) an example evaluation report which benchmarks the performance of over a dozen openly available chat-tuned language models; (7) a test specification for the benchmark.

However, their capabilities on conversational summarization remains under explored. In this work we evaluate LLMs ( 10 billion parameters) on conversational summarization and showcase their performance on various prompts. We show that the summaries generated by models depend on the instructions and the performance of LLMs vary with different instructions sometimes resulting steep drop in ROUGE scores if prompts are not selected carefully. We also evaluate the models with human evaluations and discuss the limitations of the models on conversational summarization.

Recently, researchers have made considerable improvements in dialogue systems with the progress of large language models (LLMs) such as ChatGPT and GPT-4. These LLM-based chatbots encode the potential biases while retaining disparities that can harm humans during interactions. The traditional biases investigation methods often rely on human-written test cases. However, these test cases are usually expensive and limited. In this work, we propose a first-of-its-kind method that automatically generates test cases to detect LLMs' potential gender bias. We apply our method to three well-known LLMs and find that the generated test cases effectively identify the presence of biases. To address the biases identified, we propose a mitigation strategy that uses the generated test cases as demonstrations for in-context learning to circumvent the need for parameter fine-tuning. The experimental results show that LLMs generate fairer responses with the proposed approach.

Enriching the quality of early childhood education with interactive math learning at home systems, empowered by recent advances in conversational AI technologies, is slowly becoming a reality. With this motivation, we implement a multimodal dialogue system to support play-based learning experiences at home, guiding kids to master basic math concepts. This work explores Spoken Language Understanding (SLU) pipeline within a task-oriented dialogue system developed for Kid Space, with cascading Automatic Speech Recognition (ASR) and Natural Language Understanding (NLU) components evaluated on our home deployment data with kids going through gamified math learning activities. We validate the advantages of a multi-task architecture for NLU and experiment with a diverse set of pretrained language representations for Intent Recognition and Entity Extraction tasks in the math learning domain. To recognize kids' speech in realistic home environments, we investigate several ASR systems, including the commercial Google Cloud and the latest open-source Whisper solutions with varying model sizes. We evaluate the SLU pipeline by testing our best-performing NLU models on noisy ASR output to inspect the challenges of understanding children for math learning in authentic homes.

Our work utilizes Summarization is the consolidated format for a learning of these semantic concepts as an intermediate large document and has been widely used for step from the videos. These semantic concepts many applications i.e., understanding a long meeting/event, along with the transcriptions (semantic augmentation) story summarization etc. Abstractive as input to a pre-trained summarizer model summarization is challenging in the Natural Language enrich the performance. In this work, we address Generation(NLG) domain as it requires an the problem of (i) generating semantically relevant understanding of all the salient information in annotations of a video (semantic concepts) using a the input document and rewriting logically in a fixed number of sampled frames from each video condensed manner rather than selection (extractive).

With the power of large pretrained language models, various research works have integrated knowledge into dialogue systems. The traditional techniques treat knowledge as part of the input sequence for the dialogue system, prepending a set of knowledge statements in front of dialogue history. However, such a mechanism forces knowledge sets to be concatenated in an ordered manner, making models implicitly pay imbalanced attention to the sets during training. In this paper, we first investigate how the order of the knowledge set can influence autoregressive dialogue systems' responses. We conduct experiments on two commonly used dialogue datasets with two types of transformer-based models and find that models view the input knowledge unequally. To this end, we propose a simple and novel technique to alleviate the order effect by modifying the position embeddings of knowledge input in these models. With the proposed position embedding method, the experimental results show that each knowledge statement is uniformly considered to generate responses.

Parameter-efficient methods (like Prompt or Adapters) for adapting pre-trained language models to downstream tasks have been popular recently. However, hindrances still prevent these methods from reaching their full potential. For example, two significant challenges are few-shot adaptation and cross-task generalization ability. To tackle these issues, we propose a general framework to enhance the few-shot adaptation and cross-domain generalization ability of parameter-efficient methods. In our framework, we prime the self-supervised model for parameter-efficient methods to rapidly adapt to various downstream few-shot tasks. To evaluate the authentic generalization ability of these parameter-efficient methods, we conduct experiments on a few-shot cross-domain benchmark containing 160 diverse NLP tasks. The experiment result reveals that priming by tuning PLM only with extra training tasks leads to the best performance. Also, we perform a comprehensive analysis of various parameter-efficient methods under few-shot cross-domain scenarios.