26 accepted papers
Vision-Language Models (VLMs) often suffer from visual hallucinations – generating things that are not consistent with visual inputs – and language shortcuts, where they skip the visual part and just rely on text priors. These issues arise because most post-training methods for VLMs rely on simple v…
Language models are often miscalibrated, leading to confidently incorrect answers. We introduce GRACE, a benchmark for language model calibration that incorporates comparison with human calibration. GRACE consists of question-answer pairs, in which each question contains a series of clues that gradu…
Adversarial datasets should validate AI robustness by providing samples on which humans perform well, but models do not. However, as models evolve, datasets can become obsolete. Measuring whether a dataset remains adversarial is hindered by the lack of a standardized metric for measuring adversarial…
A common use of NLP is to facilitate the understanding of large document collections, with models based on Large Language Models (LLMs) replacing probabilistic topic models. Yet the effectiveness of LLM-based approaches in real-world applications remains under explored. This study measures the knowl…
Query-focused summarization (QFS) gives a summary of documents to answer a query.Past QFS work assumes queries have one answer, ignoring debatable ones (*Is law school worth it?*).We introduce **Debatable QFS (DQFS)**, a task to create summaries that answer debatable queries via documents with oppos…
Questions help unlock information to satisfy users’ information needs. However, when the question is poorly posed, answerers (whether human or computer) may struggle to answer the question in a way that satisfies the asker, despite possibly knowing everything necessary to address the asker’s latent…
AIs can beat humans in game environments; however, how helpful those agents are to human remains understudied. We augment Cicero, a natural language agent that demonstrates superhuman performance in Diplomacy, to generate both move and message advice based on player intentions. A dozen Diplomacy gam…
Topic models and document-clustering evaluations either use automated metrics that align poorly with human preferences, or require expert labels that are intractable to scale. We design a scalable human evaluation protocol and a corresponding automated approximation that reflect practitioners’ real-…
Question answering (QA)—giving correct answers to questions—is a popular task, but we test **reverse question answering (RQA)**: for an input answer, give a question with that answer. Past work tests QA and RQA separately, but we test them jointly, comparing their difficulty, aiding benchmark design…
An increasingly common socio-technical problem is people being taken in by offers that sound “too good to be true”, where persuasion and trust shape decision-making. This paper investigates how AI can help detect these deceptive scenarios. We analyze how humans strategically deceive each other in Di…
Vision Language models (VLMs) have achieved remarkable success in video understanding tasks. Yet, a key question remains: Do they comprehend visual information or merely learn superficial mappings between visual and textual patterns? Understanding visual cues, particularly those related to physics…
Multiple choice question answering (MCQA) is popular for LLM evaluation due to its simplicity and human-like testing, but we argue for its reform. We first reveal flaws in MCQA’s format, as it struggles to: 1) test generation/subjectivity; 2) match LLM use cases; and 3) fully test knowledge. We inst…
LLMs are aligned to follow input instructions by learning which of two responses users prefer for a prompt. However, such preference data do not convey *why* users prefer responses that are chosen or rejected, so LLMs trained on these datasets cannot tailor responses to varied user needs. To surface…
Keyword mnemonics are memorable explanations that link new terms to simpler keywords.Prior work generates mnemonics for students, but they do not train models using mnemonics students prefer and aid learning.We build SMART, a mnemonic generator trained on feedback from real students learning new ter…
Large vision-language models (LVLMs) are prone to hallucinations, where certain contextual cues in an image can trigger the language module to produce overconfident and incorrect reasoning about abnormal or hypothetical objects. While some benchmarks have been developed to investigate LVLM hallucina…
Recent advancements of large language models (LLMs)have led to claims of AI surpassing humansin natural language processing NLP tasks such as textual understanding and reasoning.%This work investigates these assertions by introducingCAIMIRA, a novel framework rooted in item response theory IRTthat e…
Flashcard schedulers rely on 1) *student models* to predict the flashcards a student knows; and 2) *teaching policies* to pick which cards to show next via these predictions.Prior student models, however, just use study data like the student’s past responses, ignoring the text on cards. We propose *…
The boardgame Diplomacy is a challenging setting for communicative and cooperative artificial intelligence. The most prominent communicative Diplomacy AI, Cicero, has excellent strategic abilities, exceeding human players. However, the best Diplomacy players master communication, not just tactics, w…
Question answering (QA) can only make progress if we know if an answer is correct, but current answer correctness (AC) metrics struggle with verbose, free-form answers from large language models (LLMs). There are two challenges with current short-form QA evaluations: a lack of diverse styles of eval…
Automating the creation of scientific diagrams from academic papers can significantly streamline the development of tutorials, presentations, and posters, thereby saving time and accelerating the process. Current text-to-image models (Rombach et al., 2022a; Belouadi et al., 2023) struggle with gener…
Training question-answering QA and information retrieval systems for web queries require large, expensive datasets that are difficult to annotate and time-consuming to gather. Moreover, while natural datasets of information-seeking questions are often prone to ambiguity or ill-formed, there are trov…
Translations help people understand content written in another language. However, even correct literal translations do not fulfill that goal when people lack the necessary background to understand them. Professional translators incorporate explicitations to explain the missing context by considering…
While recent large language models (LLMs) improve on various question answering (QA) datasets, it remains difficult for a single model to generalize across question types that require distinct reasoning abilities. We provide empirical evidence that state-of-the-art LLMs suffer from poor generalizabi…
Large Language Models (LLMs) are increasingly being deployed in interactive contexts that involve direct user engagement, such as chatbots and writing assistants. These deployments are increasingly plagued by prompt injection and jailbreaking (collectively, prompt hacking), in which models are manip…
Large language models (LLMs) show impressive abilities via few-shot prompting. Commercialized APIs such as OpenAI GPT-3 further increase their use in real-world language applications. However, the crucial problem of how to improve the reliability of GPT-3 is still under-explored. While reliability i…
Topic model evaluation, like evaluation of other unsupervised methods, can be contentious. However, the field has coalesced around automated estimates of topic coherence, which rely on the frequency of word co-occurrences in a reference corpus. Contemporary neural topic models surpass classical ones…