Roee Aharoni
Roee Aharoni is a Research Scientist at Google Tel Aviv, working on natural language processing.
Research Areas
Authored Publications
Sort By
(D)RAGged Into a Conflict: Detecting and Addressing Conflicting Sources in Retrieval-Augmented LLMs
Arie Cattan
Alon Jacovi
Ori Ram
Eran Ofek
2025
Preview abstract
Retrieval Augmented Generation (RAG) is a commonly used approach for enhancing LLMs with relevant and up-to-date information. However, the retrieved sources can often bring conflicting information and it is not clear how models address such discrepancies. In this work, we first point out that knowledge conflicts stem from various reasons and thus require tailored solutions in order to better align model responses to human preferences. To that end, we introduce a novel taxonomy of knowledge conflicts in RAG and define the desired model’s behavior for each category. Additionally, we construct a high-quality benchmark by asking two expert annotators to identify the conflict type within realistic RAG instances, each comprising a query and its associated search results. Finally, we conduct extensive experiments and show that explicitly informing LLMs about the potential conflict category significantly improves the quality and appropriateness of the responses. Yet, there is still a vast room for improvement. Taken together, our work highlights the importance of evaluating RAG systems not only on factual accuracy but also on their ability to manage and resolve knowledge conflicts effectively.
View details
Preview abstract
Recent work suggested utilizing inference compute, showing that scaling of number of samples consistently improves the fractions of problems solved by any attempt, namely the coverage. In this work, we suggest that inference scaling gains should be compared with proper baselines, as some datasets become degenerate when allowing a large number of attempts. We focus on two domains - mathematical reasoning and factual knowledge, showing that for the MATH and Entity Questions datasets, informed answer enumeration obtains similar or even better results than repeated model sampling, with a much lower sample budget. While we believe that inference scaling is a promising approach for unlocking the potential of language models, we recommend carefully selecting models and datasets when applying this method. Otherwise, the results of inference scaling should be interpreted with caution.
View details
Preview abstract
Factual questions typically can be answered correctly at different levels of granularity. For example, both “August 4, 1961” and “1961” are correct answers to the question “When was Barack Obama born?”. Standard question answering (QA) evaluation protocols, however, do not explicitly take this into account and compare a predicted answer against answers of a single granularity level.
In this work, we propose GRANOLA QA, a novel evaluation setting where a predicted answer is evaluated in terms of accuracy and informativeness against a set of multi-granularity answers. We present a simple methodology for enriching existing datasets with multi-granularity answers, and create GRANOLA-EQ, a multi-granularity version of the EntityQuestions dataset.
We evaluate a range of decoding methods on GRANOLA-EQ, including a new algorithm, called Decoding with Response Aggregation (DRAG), that is geared towards aligning the response granularity with the model's uncertainty. Our experiments show that large language models with standard decoding tend to generate specific answers, which are often incorrect. In contrast, when evaluated on multi-granularity answers, DRAG yields a nearly 20 point increase in accuracy on average, which further increases for rare entities.Overall, this reveals that standard evaluation and decoding schemes may significantly underestimate the knowledge encapsulated in LMs.
View details
Preview abstract
As instruction-tuned large language models (LLMs) gain global adoption, their ability to follow instructions in multiple languages becomes increasingly crucial. In this work, we investigate how multilinguality during instruction tuning of a multilingual LLM affects instruction-following across languages from the pre-training corpus. We first show that many languages transfer some instruction-following capabilities to other languages from even monolingual tuning. Furthermore, we find that only 40 multilingual examples integrated in an English tuning set substantially improve multilingual instruction-following, both in seen and unseen languages during tuning. In general, we observe that models tuned on multilingual mixtures exhibit comparable or superior performance in multiple languages compared to monolingually tuned models, despite training on 10x fewer examples in those languages. Finally, we find that diversifying the instruction tuning set with even just 2-4 languages significantly improves cross-lingual generalization. Our results suggest that building massively multilingual instruction-tuned models can be done with only a very small set of multilingual instruction-responses.
View details
A Chain-of-Thought Is as Strong as Its Weakest Link: A Benchmark for Verifiers of Reasoning Chains
Alon Jacovi
Or Honovich
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (2024), pp. 4615–4634
Preview abstract
Prompting language models to provide step-by-step answers (e.g., “Chain-of-Thought”) is the prominent approach for complex reasoning tasks, where more accurate reasoning chains typically improve downstream task performance. Recent literature discusses automatic methods to verify reasoning to evaluate and improve their correctness. However, no fine-grained step-level datasets are available to enable thorough evaluation of such verification methods, hindering progress in this direction. We introduce REVEAL: Reasoning Verification Evaluation, a dataset to benchmark automatic verifiers of complex Chain-of-Thought reasoning in open-domain question-answering settings. REVEAL includes comprehensive labels for the relevance, attribution to evidence passages, and logical correctness of each reasoning step in a language model’s answer, across a variety of datasets and state-of-the-art language models. Evaluation on REVEAL shows that verifiers struggle at verifying reasoning chains — in particular, verifying logical correctness and detecting contradictions. Available at https://reveal-dataset.github.io/.
View details
Preview abstract
We posit that large language models (LLMs) should be capable of expressing their intrinsic uncertainty in natural language. For example, if the LLM is equally likely to output two contradicting answers to the same question, then its generated response should reflect this uncertainty by hedging its answer (e.g., "I'm not sure, but I think..."). We formalize faithful response uncertainty based on the gap between the model's intrinsic confidence in the assertions it makes and the decisiveness by which they are conveyed. This example-level metric reliably indicates whether the model reflects its uncertainty, as it penalizes both excessive and insufficient hedging. We evaluate a variety of aligned LLMs at faithfully communicating uncertainty on several knowledge-intensive question answering tasks. Our results provide strong evidence that modern LLMs are poor at faithfully conveying their uncertainty, and that better alignment is necessary to improve their trustworthiness.
View details
A Comprehensive Evaluation of Tool-Assisted Generation Strategies
Alon Jacovi
Findings of EMNLP (2023)
Preview abstract
A growing area of research investigates augmenting language models with tools (e.g., search engines, calculators) to overcome their shortcomings (e.g., missing or incorrect knowledge, incorrect logical inferences). Various few-shot tool-usage strategies have been proposed. However, there is no systematic and fair comparison across different strategies, or between these strategies and strong baselines that do not leverage tools. We conduct an extensive empirical analysis, finding that (1) across various datasets, example difficulty levels, and models, strong no-tool baselines are competitive to tool-assisted strategies, implying that effectively using tools with in-context demonstrations is a difficult unsolved problem; (2) for knowledge-retrieval tasks, strategies that *refine* incorrect outputs with tools outperform strategies that retrieve relevant information *ahead of* or *during generation*; (3) tool-assisted strategies are expensive in the number of tokens they require to work -- incurring additional costs by orders of magnitude -- which does not translate into significant improvement in performance. Overall, our findings suggest that few-shot tool integration is still an open challenge, emphasizing the need for comprehensive evaluations of future strategies to accurately assess their *benefits* and *costs*.
View details
q2d: Automatic Dialog Generation to Improve Models' Query Generation
Enav Weinreb
Ido Hakimi
Shlomi Cohen-Ganor
Yoad Lewenberg
EMNLP 2023 (2023)
Preview abstract
We propose q2d: an automatic data generation pipeline that generates information-seeking dialogues based on questions. We apply our method to create conversational versions of questions answering datasets, which we release as a new dataset. We use this data to improve query generation models, which communicate with an external search APIs to generate factual responses. Unlike previous approaches, which relied on human annotators, our method allows to automatically generate labeled dialogues with better control and scale.
In experiments, we demonstrate that: (1) Models trained on our synthetic data produce results comparable to those trained on natural data; (2) Our generated datasets are effective as a benchmark and as a training signal that generalizes to human-annotated test sets.
We also provide an extensive analysis of the quality and factuality of the generated datasets. Our studies indicate that our automatic dialogue generation pipeline is effective at improving query generation and factuality.
View details
Preview abstract
Factual consistency evaluation is often conducted using Natural Language Inference (NLI) models, yet these models exhibit limited success in evaluating summaries. Previous work improved such models with synthetic training data. However, the data is typically based on perturbed human-written summaries, which often differ in their characteristics from real model-generated summaries and have limited coverage of possible factual errors. Alternatively, large language models (LLMs) have recently shown promising results in directly evaluating generative tasks, but are too computationally expensive for practical use. Motivated by these limitations, we introduce TrueTeacher, a method for generating synthetic data by annotating diverse model-generated summaries using a LLM. Unlike prior work, TrueTeacher does not rely on human-written summaries, and is multilingual by nature. Experiments on the TRUE benchmark show that a student model trained using our data, substantially outperforms both the state-of-the-art model with similar capacity, and the LLM teacher. In a systematic study, we compare TrueTeacher to existing synthetic data generation methods and demonstrate its superiority and robustness to domain-shift. We also show that our method generalizes to multilingual scenarios using the mFACE dataset. Finally, we release a large-scale synthetic dataset with 1.4M examples generated using TrueTeacher.
View details
SEAHORSE: A Dataset of Summaries Annotated with Human Ratings in Six Languages
Elizabeth Clark
Shruti Rijhwani
Sebastian Gehrmann
Aditya Siddhant
EMNLP 2023, Association for Computational Linguistics (2023)
Preview abstract
We introduce Seahorse (SummariEs Annotated with Human Ratings in Six languagEs), a dataset of 96K summaries with ratings along 6 dimensions (comprehensibility, repetition, grammar, attribution, main idea(s), and conciseness). The summaries are generated from 8 different models, conditioned on source text from 4 datasets in 6 languages (German, English, Spanish, Russian, Turkish, and Vietnamese). We release the annotated summaries as a resource for developing better summarization models and automatic metrics. We present an analysis of the dataset's composition and quality, and we demonstrate the potential of this dataset for building better summarization metrics, showing that metrics finetuned with Seahorse data outperform baseline metrics.
View details
