CONSCENDI: A Contrastive and Scenario-Guided Distillation Approach to Guardrail Models for Virtual Assistants

• URL: http://arxiv.org/abs/2304.14364v2
• Date: Wed, 3 Apr 2024 23:57:12 GMT
• Title: CONSCENDI: A Contrastive and Scenario-Guided Distillation Approach to Guardrail Models for Virtual Assistants
• Authors: Albert Yu Sun, Varun Nair, Elliot Schumacher, Anitha Kannan,
• Abstract summary: A major challenge in deploying large language models (LLMs) is ensuring they operate within what is admissible for the task. We propose CONSCENDI to exhaustively generate training data with two key components: scenario-augmented generation and contrastive training examples. We find that CONSCENDI results in guardrail models that improve over baselines in multiple dialogue domains.
• Score: 9.912132935716116
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: A wave of new task-based virtual assistants has been fueled by increasingly powerful large language models (LLMs), such as GPT-4 (OpenAI, 2023). A major challenge in deploying LLM-based virtual conversational assistants in real world settings is ensuring they operate within what is admissible for the task. To overcome this challenge, the designers of these virtual assistants rely on an independent guardrail system that verifies the virtual assistant's output aligns with the constraints required for the task. However, relying on commonly used, prompt-based guardrails can be difficult to engineer correctly and comprehensively. To address these challenges, we propose CONSCENDI. We use CONSCENDI to exhaustively generate training data with two key LLM-powered components: scenario-augmented generation and contrastive training examples. When generating conversational data, we generate a set of rule-breaking scenarios, which enumerate a diverse set of high-level ways a rule can be violated. This scenario-guided approach produces a diverse training set and provides chatbot designers greater control. To generate contrastive examples, we prompt the LLM to alter conversations with violations into acceptable conversations to enable fine-grained distinctions. We then use this data, generated by CONSCENDI, to train a smaller model. We find that CONSCENDI results in guardrail models that improve over baselines in multiple dialogue domains.

Related papers

• Generative Context Distillation [48.91617280112579]
  Generative Context Distillation (GCD) is a lightweight prompt internalization method that employs a joint training approach. We demonstrate that our approach effectively internalizes complex prompts across various agent-based application scenarios.
  arXiv  Detail & Related papers  (2024-11-24T17:32:20Z)
• A Flexible Large Language Models Guardrail Development Methodology Applied to Off-Topic Prompt Detection [0.0]
  Large Language Models are prone to off-topic misuse, where users may prompt these models to perform tasks beyond their intended scope. Current guardrails suffer from high false-positive rates, limited adaptability, and the impracticality of requiring real-world data that is not available in pre-production. This paper introduces a flexible, data-free guardrail development methodology that addresses these challenges.
  arXiv  Detail & Related papers  (2024-11-20T00:31:23Z)
• Many Hands Make Light Work: Task-Oriented Dialogue System with Module-Based Mixture-of-Experts [9.129081545049992]
  Task-oriented dialogue systems have greatly benefited from pre-trained language models (PLMs) We propose Soft Mixture-of-Expert Task-Oriented Dialogue system (SMETOD) SMETOD leverages an ensemble of Mixture-of-Experts (MoEs) to excel at subproblems and generate specialized outputs for task-oriented dialogues. We extensively evaluate our model on three benchmark functionalities: intent prediction, dialogue state tracking, and dialogue response generation.
  arXiv  Detail & Related papers  (2024-05-16T01:02:09Z)
• WavLLM: Towards Robust and Adaptive Speech Large Language Model [93.0773293897888]
  We introduce WavLLM, a robust and adaptive speech large language model with dual encoders, and a prompt-aware LoRA weight adapter. We validate the proposed model on universal speech benchmarks including tasks such as ASR, ST, SV, ER, and also apply it to specialized datasets like Gaokao English listening comprehension set for SQA, and speech Chain-of-Thought (CoT) evaluation set.
  arXiv  Detail & Related papers  (2024-03-31T12:01:32Z)
• Self-regulating Prompts: Foundational Model Adaptation without Forgetting [112.66832145320434]
  We introduce a self-regularization framework for prompting called PromptSRC. PromptSRC guides the prompts to optimize for both task-specific and task-agnostic general representations.
  arXiv  Detail & Related papers  (2023-07-13T17:59:35Z)
• Task-Optimized Adapters for an End-to-End Task-Oriented Dialogue System [0.0]
  We propose an End-to-end TOD system with Task-d Adapters which learn independently per task, adding only small number of parameters after fixed layers of pre-trained network. Our method is a model-agnostic approach and does not require prompt-tuning as only input data without a prompt.
  arXiv  Detail & Related papers  (2023-05-04T00:17:49Z)
• Guiding Large Language Models via Directional Stimulus Prompting [114.84930073977672]
  We introduce Directional Stimulus Prompting, a novel framework for guiding black-box large language models (LLMs) toward specific desired outputs. Instead of directly adjusting LLMs, our method employs a small tunable policy model to generate an auxiliary directional stimulus prompt for each input instance.
  arXiv  Detail & Related papers  (2023-02-22T17:44:15Z)
• Stabilized In-Context Learning with Pre-trained Language Models for Few Shot Dialogue State Tracking [57.92608483099916]
  Large pre-trained language models (PLMs) have shown impressive unaided performance across many NLP tasks. For more complex tasks such as dialogue state tracking (DST), designing prompts that reliably convey the desired intent is nontrivial. We introduce a saliency model to limit dialogue text length, allowing us to include more exemplars per query.
  arXiv  Detail & Related papers  (2023-02-12T15:05:10Z)
• Making Pre-trained Language Models End-to-end Few-shot Learners with Contrastive Prompt Tuning [41.15017636192417]
  We present CP-Tuning, the first end-to-end Contrastive Prompt Tuning framework for fine-tuning Language Models. It is integrated with the task-invariant continuous prompt encoding technique with fully trainable prompt parameters. Experiments over a variety of language understanding tasks used in IR systems and different PLMs show that CP-Tuning outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2022-04-01T02:24:24Z)
• CINS: Comprehensive Instruction for Few-shot Learning in Task-oriented Dialog Systems [56.302581679816775]
  This paper proposes Comprehensive Instruction (CINS) that exploits PLMs with task-specific instructions. We design a schema (definition, constraint, prompt) of instructions and their customized realizations for three important downstream tasks in ToD. Experiments are conducted on these ToD tasks in realistic few-shot learning scenarios with small validation data.
  arXiv  Detail & Related papers  (2021-09-10T03:23:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.