Despite growing interest in automated hate speech detection, most existing approaches overlook the linguistic diversity of online content. Multilingual instruction-tuned large language models such as LLaMA, Aya, Qwen, and BloomZ offer promising capabilities across languages, but their effectiveness in identifying hate speech through zero-shot and few-shot prompting remains underexplored. This work evaluates LLM prompting-based detection across eight non-English languages, utilizing several prompting techniques and comparing them to fine-tuned encoder models. We show that while zero-shot and few-shot prompting lag behind fine-tuned encoder models on most of the real-world evaluation sets, they achieve better generalization on functional tests for hate speech detection. Our study also reveals that prompt design plays a critical role, with each language often requiring customized prompting techniques to maximize performance.

One of the major aspects contributing to the striking performance of large language models (LLMs) is the vast amount of factual knowledge accumulated during pre-training. Yet, many LLMs suffer from self-inconsistency, which raises doubts about their trustworthiness and reliability. In this paper, we focus on entity type ambiguity and analyze current state-of-the-art LLMs for their proficiency and consistency in applying their factual knowledge when prompted for entities under ambiguity. To do so, we propose an evaluation protocol that disentangles knowing from applying knowledge, and test state-of-the-art LLMs on 49 entities. Our experiments reveal that LLMs perform poorly with ambiguous prompts, achieving only 80% accuracy. Our results further demonstrate systematic discrepancies in LLM behavior and their failure to consistently apply information, indicating that the models can exhibit knowledge without being able to utilize it, significant biases for preferred readings, as well as self inconsistencies. Our study highlights the importance of handling entity ambiguity in future for more trustworthy LLMs

We propose MMLU-SR, a novel dataset designed to measure the true comprehension abilities of Large Language Models (LLMs) by challenging their performance in question-answering tasks with modified terms. We reasoned that an agent that ``truly'' understands a concept can still evaluate it when key terms are replaced by suitably defined alternate terms, and sought to differentiate such comprehension from mere text replacement. In our study, we modified standardized test questions by replacing a key term with a dummy word along with its definition. The key term could be in the context of questions, answers, or both questions and answers. Notwithstanding the high scores achieved by recent popular LLMs on the MMLU leaderboard, we found a substantial reduction in model performance after such replacement, suggesting poor comprehension. This new benchmark provides a rigorous benchmark for testing true model comprehension, and poses a challenge to the broader scientific community.

The increasing threat of disinformation calls for automating parts of the fact-checking pipeline. Identifying text segments requiring fact-checking is known as claim detection (CD) and claim check-worthiness detection (CW), the latter incorporating complex domain-specific criteria of worthiness and often framed as a ranking task. Zero- and few-shot LLM prompting is an attractive option for both tasks, as it bypasses the need for labeled datasets and allows verbalized claim and worthiness criteria to be directly used for prompting. We evaluate the LLMs' predictive and calibration accuracy on five CD/CW datasets from diverse domains, each utilizing a different worthiness criterion. We investigate two key aspects: (1) how best to distill factuality and worthiness criteria into a prompt and (2) what amount of context to provide for each claim. To this end, we experiment with varying the level of prompt verbosity and the amount of contextual information provided to the model. Our results show that optimal prompt verbosity is domain-dependent, adding context does not improve performance, and confidence scores can be directly used to produce reliable check-worthiness rankings.

This paper presents GRASP, a novel benchmark to evaluate the language grounding and physical understanding capabilities of video-based multimodal large language models (LLMs). This evaluation is accomplished via a two-tier approach leveraging Unity simulations. The first level tests for language grounding by assessing a model's ability to relate simple textual descriptions with visual information. The second level evaluates the model's understanding of "Intuitive Physics" principles, such as object permanence and continuity. In addition to releasing the benchmark, we use it to evaluate several state-of-the-art multimodal LLMs. Our evaluation reveals significant shortcomings in the language grounding and intuitive physics capabilities of these models. Although they exhibit at least some grounding capabilities, particularly for colors and shapes, these capabilities depend heavily on the prompting strategy. At the same time, all models perform below or at the chance level of 50% in the Intuitive Physics tests, while human subjects are on average 80% correct. These identified limitations underline the importance of using benchmarks like GRASP to monitor the progress of future models in developing these competencies.

Language models can be prompted to reason through problems in a manner that significantly improves performance. However, \textit{why} such prompting improves performance is unclear. Recent work showed that using logically \textit{invalid} Chain-of-Thought (CoT) prompting improves performance almost as much as logically \textit{valid} CoT prompting, and that editing CoT prompts to replace problem-specific information with abstract information or out-of-distribution information typically doesn't harm performance. Critics have responded that these findings are based on too few and too easily solved tasks to draw meaningful conclusions. To resolve this dispute, we test whether logically invalid CoT prompts offer the same level of performance gains as logically valid prompts on the hardest tasks in the BIG-Bench benchmark, termed BIG-Bench Hard (BBH). We find that the logically \textit{invalid} reasoning prompts do indeed achieve similar performance gains on BBH tasks as logically valid reasoning prompts. We also discover that some CoT prompts used by previous works contain logical errors. This suggests that covariates beyond logically valid reasoning are responsible for performance improvements.

Question answering over knowledge bases (KBs) aims to answer natural language questions with factual information such as entities and relations in KBs. Previous methods either generate logical forms that can be executed over KBs to obtain final answers or predict answers directly. Empirical results show that the former often produces more accurate answers, but it suffers from non-execution issues due to potential syntactic and semantic errors in the generated logical forms. AF that jointly generates both logical forms and direct answers, and then combines the merits of them to get the final answers. AF is based on simple free-text retrieval without relying on any entity linking tools -- this simplification eases its adaptation to different datasets. AF achieves new stateof-the-art accuracy on WebQSP, FreebaseQA, and GrailQA benchmarks, while getting competitive results on the ComplexWebQuestions benchmark. Knowledge Bases Question Answering (KBQA) aims to answer natural language questions based on knowledge from KBs such as DBpedia (Auer et al., 2007), Freebase (Bollacker et al., 2008) or Wikidata (Vrandečić & Krötzsch, 2014). Existing methods can be divided into two categories. One category is based on semantic parsing, where models first parse the input question into a logical form (e.g., SPARQL (hommeaux, 2011) or S-expression (Gu et al., 2021)) then execute the logical form against knowledge bases to obtain the final answers (Das et al., 2021; Gu et al., 2021; Ye et al., 2022). They either classify the entities in KB to decide which are the answers (Sun et al., 2019) or generate the answers using a sequence-to-sequence framework (Saxena et al., 2022; Oğuz et al., 2022). Previous empirical results (Ye et al., 2022; Das et al., 2021; Gu et al., 2022) show that the semantic parsing based methods can produce more accurate answers over benchmark datasets. However, due to the syntax and semantic restrictions, the output logical forms can often be non-executable and thus would not produce any answers. On the other hand, direct-answer-prediction methods can guarantee to generate output answers, albeit their answer accuracy is usually not as good as semantic parsing based methods, especially over complex questions which require multi-hop reasoning (Talmor & Berant, 2018).