UniIR: Training and Benchmarking Universal Multimodal Information Retrievers

Existing information retrieval (IR) models often assume a homogeneous format, limiting their applicability to diverse user needs, such as searching for images with text descriptions, searching for a news article with a headline image, or finding a similar photo with a query image. To approach such different information-seeking demands, we introduce UniIR, a unified instruction-guided multimodal retriever capable of handling eight distinct retrieval tasks across modalities. UniIR, a single retrieval system jointly trained on ten diverse multimodal-IR datasets, interprets user instructions to execute various retrieval tasks, demonstrating robust performance across existing datasets and zero-shot generalization to new tasks. Our experiments highlight that multi-task training and instruction tuning are keys to UniIR's generalization ability. Additionally, we construct the M-BEIR, a multimodal retrieval benchmark with comprehensive results, to standardize the evaluation of universal multimodal information retrieval.

Reducing Privacy Risks in Online Self-Disclosures with Language Models

Self-disclosure, while being common and rewarding in social media interaction, also poses privacy risks. In this paper, we take the initiative to protect the user-side privacy associated with online self-disclosure through identification and abstraction. We develop a taxonomy of 19 self-disclosure categories, and curate a large corpus consisting of 4.8K annotated disclosure spans. We then fine-tune a language model for identification, achieving over 75% in Token F$_1$. We further conduct a HCI user study, with 82\% of participants viewing the model positively, highlighting its real world applicability. Motivated by the user feedback, we introduce the task of self-disclosure abstraction. We experiment with both one-span abstraction and three-span abstraction settings, and explore multiple fine-tuning strategies. Our best model can generate diverse abstractions that moderately reduce privacy risks while maintaining high utility according to human evaluation.

Tensor Trust: Interpretable Prompt Injection Attacks from an Online Game

While Large Language Models (LLMs) are increasingly being used in real-world applications, they remain vulnerable to prompt injection attacks: malicious third party prompts that subvert the intent of the system designer. To help researchers study this problem, we present a dataset of over 126,000 prompt injection attacks and 46,000 prompt-based "defenses" against prompt injection, all created by players of an online game called Tensor Trust. To the best of our knowledge, this is currently the largest dataset of human-generated adversarial examples for instruction-following LLMs. The attacks in our dataset have a lot of easily interpretable stucture, and shed light on the weaknesses of LLMs. We also use the dataset to create a benchmark for resistance to two types of prompt injection, which we refer to as prompt extraction and prompt hijacking. Our benchmark results show that many models are vulnerable to the attack strategies in the Tensor Trust dataset. Furthermore, we show that some attack strategies from the dataset generalize to deployed LLM-based applications, even though they have a very different set of constraints to the game. We release all data and source code at https://tensortrust.ai/paper

Can Language Models be Instructed to Protect Personal Information?

Large multimodal language models have proven transformative in numerous applications. However, these models have been shown to memorize and leak pre-training data, raising serious user privacy and information security concerns. While data leaks should be prevented, it is also crucial to examine the trade-off between the privacy protection and model utility of proposed approaches. In this paper, we introduce PrivQA -- a multimodal benchmark to assess this privacy/utility trade-off when a model is instructed to protect specific categories of personal information in a simulated scenario. We also propose a technique to iteratively self-moderate responses, which significantly improves privacy. However, through a series of red-teaming experiments, we find that adversaries can also easily circumvent these protections with simple jailbreaking methods through textual and/or image inputs. We believe PrivQA has the potential to support the development of new models with improved privacy protections, as well as the adversarial robustness of these protections. We release the entire PrivQA dataset at https://llm-access-control.github.io/.

Self-Specialization: Uncovering Latent Expertise within Large Language Models

Recent works have demonstrated the effectiveness of self-alignment in which a large language model is, by itself, aligned to follow general instructions through the automatic generation of instructional data using a handful of human-written seeds. Instead of general alignment, in this work, we focus on self-alignment for expert domain specialization (e.g., biomedicine), discovering it to be very effective for improving zero-shot and few-shot performance in target domains of interest. As a preliminary, we first present the benchmark results of existing aligned models within a specialized domain, which reveals the marginal effect that "generic" instruction-following training has on downstream expert domains' performance. To remedy this, we explore self-specialization that leverages domain-specific unlabelled data and a few labeled seeds for the self-alignment process. When augmented with retrieval to reduce hallucination and enhance concurrency of the alignment, self-specialization offers an effective (and efficient) way of "carving out" an expert model out of a "generalist", pre-trained LLM where different domains of expertise are originally combined in a form of "superposition". Our experimental results on a biomedical domain show that our self-specialized model (30B) outperforms its base model, MPT-30B by a large margin and even surpasses larger popular models based on LLaMA-65B, highlighting its potential and practicality for specialization, especially considering its efficiency in terms of data and parameters.

Improved Instruction Ordering in Recipe-Grounded Conversation

In this paper, we study the task of instructional dialogue and focus on the cooking domain. Analyzing the generated output of the GPT-J model, we reveal that the primary challenge for a recipe-grounded dialog system is how to provide the instructions in the correct order. We hypothesize that this is due to the model's lack of understanding of user intent and inability to track the instruction state (i.e., which step was last instructed). Therefore, we propose to explore two auxiliary subtasks, namely User Intent Detection and Instruction State Tracking, to support Response Generation with improved instruction grounding. Experimenting with our newly collected dataset, ChattyChef, shows that incorporating user intent and instruction state information helps the response generation model mitigate the incorrect order issue. Furthermore, to investigate whether ChatGPT has completely solved this task, we analyze its outputs and find that it also makes mistakes (10.7% of the responses), about half of which are out-of-order instructions. We will release ChattyChef to facilitate further research in this area at: https://github.com/octaviaguo/ChattyChef.

Better Low-Resource Entity Recognition Through Translation and Annotation Fusion

Pre-trained multilingual language models have enabled significant advancements in cross-lingual transfer. However, these models often exhibit a performance disparity when transferring from high-resource languages to low-resource languages, especially for languages that are underrepresented or not in the pre-training data. Motivated by the superior performance of these models on high-resource languages compared to low-resource languages, we introduce a Translation-and-fusion framework, which translates low-resource language text into a high-resource language for annotation using fully supervised models before fusing the annotations back into the low-resource language. Based on this framework, we present TransFusion, a model trained to fuse predictions from a high-resource language to make robust predictions on low-resource languages. We evaluate our methods on two low-resource named entity recognition (NER) datasets, MasakhaNER2.0 and LORELEI NER, covering 25 languages, and show consistent improvement up to +16 F$_1$ over English fine-tuning systems, achieving state-of-the-art performance compared to Translate-train systems. Our analysis depicts the unique advantages of the TransFusion method which is robust to translation errors and source language prediction errors, and complimentary to adapted multilingual language models.

Instruction Tuning with Lexicons for Zero-Shot Style Classification

Style is used to convey authors' intentions and attitudes. Despite the success of large pre-trained language models on style classification, prior work relies on fine-tuning with labeled examples. Prompting large language models to classify style without fine-tuning is challenging because language styles can be difficult to define. In this study, we investigate the effectiveness of style lexicons as a means for instructing language models how to identify new styles that are unseen during training. Our experiments show that lexicon-based instructions improve transfer zero-shot performance significantly. We will release our code and data.

Schema-Driven Information Extraction from Heterogeneous Tables

In this paper, we explore the question of whether language models (LLMs) can support cost-efficient information extraction from complex tables. We introduce schema-driven information extraction, a new task that uses LLMs to transform tabular data into structured records following a human-authored schema. To assess various LLM's capabilities on this task, we develop a benchmark composed of tables from three diverse domains: machine learning papers, chemistry tables, and webpages. Accompanying the benchmark, we present InstrucTE, a table extraction method based on instruction-tuned LLMs. This method necessitates only a human-constructed extraction schema, and incorporates an error-recovery strategy. Notably, InstrucTE demonstrates competitive performance without task-specific labels, achieving an F1 score ranging from 72.3 to 95.7. Moreover, we validate the feasibility of distilling more compact table extraction models to minimize extraction costs and reduce API reliance. This study paves the way for the future development of instruction-following models for cost-efficient table extraction.

Are Large Language Models Robust Zero-shot Coreference Resolvers?

Recent progress in domain adaptation for coreference resolution relies on continued training using annotated data from target domains. At the same time, pre-trained large language models (LMs) have exhibited strong zero- and few-shot learning abilities across a wide range of NLP tasks including pronoun resolution. While this demonstrates evidence of coreference ability, previous work has mostly studied this ability using simple sentence-level datasets such as the Winograd Schema Challenge. In this work, we assess the feasibility of zero-shot learning for coreference resolution by evaluating instruction-tuned language models on more difficult, linguistically-complex coreference benchmarks (e.g., CoNLL-2012). We demonstrate that zero-shot prompting outperforms current unsupervised coreference systems. Further investigations reveal the robust zero-shot generalization ability of instruction-tuned LMs across a wide range of domains, languages, and time periods, as well as a strong reliance on high-quality mention detection systems.