arXiv subjects

Marzena Karpinska

Publications and source records attributed to Marzena Karpinska.

At least 19 recordsLinked to original sources

arXiv2026

OverThink: Slowdown Attacks on Reasoning LLMs

A reasoning language model (RLM) generates costly reasoning tokens, often hidden from the users, that help it excel at many tasks. Our Overthink attack targets RLM-based applications (such as chatbots or coding agents) that rely on external context by forcing these models to generate substantially more reasoning tokens while still producing contextually correct answers. An adversary conducts the attack by injecting decoy reasoning problems into available content, optimized to elicit a large number of tokens. We craft decoy challenges (using Markov decision processes, language translation, or graphic comprehension) that appear benign individually yet have an adversarial impact when inserted in the context, allowing them to easily evade safety filters. We evaluate Overthink on proprietary and open-source reasoning models across the FreshQA, SQuAD, and MuSR datasets, where we observe 13x, 46x, and 12x increases, respectively. We also explore multimodal attacks using images, which cause up to a 2.7x increase in reasoning, as well as attacks on coding agents by injecting decoys into skills, README files, and code, resulting in up to a 17x increase. We explore several defenses and evaluate their efficacy against different attack strategies, highlighting that defending against Overthink is nontrivial. Finally, we show that newer generations of RLMs, while showing a drastic increase in per-token cost, also exhibit up to a 2.3x increase in reasoning tokens, leaving them more vulnerable to Overthink attacks.

cs.LG↗

arXiv2026

AI translation of literary texts is "fine", but readers still prefer human translations

AI translation of literary works is increasingly common. While the content may be rendered adequately, we do not know enough about how readers experience it in terms of immersiveness and literary effect-aspects poorly captured by automatic metrics or human evaluation targeting fluency and adequacy. We ask 15 avid readers to compare recently published human translations (HT) to machine translations (MT) generated with an agentic language model-based pipeline, for 15 recent novels in French, Polish, and Japanese translated into English. Readers evaluated approximately 8K-word excerpts in two conditions: immersive reading of the whole excerpt (30 comparisons) and close reading of 386 aligned HT-MT chunk pairs (772 comparisons), with two readers per book. Overall, readers find MT "fine", but prefer HT (descriptively at the excerpt level, 19/30, and significantly at the chunk level, 522/772) for its ease, clarity, and immersive nature. Readers' highlights show that MT's quality varies more within one book than HT's does. Crucially, readers cannot reliably tell the two apart (17/30 guess correctly) and tend to prefer the version they believe to be human. Automatic metrics, including LLM-as-a-judge approaches, fail to recover reader preferences and favor MT. We release LAIT (Literary AI Translation), a reader-centered evaluation dataset with 1K reader comments, 2K judgments and preference ratings, and 7.2K span-level annotations, along with our evaluation protocol and supporting interface.

cs.CL↗

arXiv2026

AI use in American newspapers is widespread, uneven, and rarely disclosed

AI is rapidly transforming journalism, but the extent of its use in published newspaper articles remains unclear. We address this gap by auditing a large-scale dataset of 186K articles from online editions of 1.5K American newspapers published in the summer of 2025. Using Pangram, a state-of-the-art AI detector, we discover that approximately 9% of newly-published articles are either partially or fully AI-generated. This AI use is unevenly distributed, appearing more frequently in smaller, local outlets, in specific topics such as weather and technology, and within certain ownership groups. We also analyze 45K opinion pieces from Washington Post, New York Times, and Wall Street Journal, finding that they are 6.4 times more likely to contain AI-generated content than news articles from the same publications, with many AI-flagged op-eds authored by prominent public figures. Despite this prevalence, we find that AI use is rarely disclosed: a manual audit of 100 AI-flagged articles found only five disclosures of AI use. Overall, our audit highlights the immediate need for greater transparency and updated editorial standards regarding the use of AI in journalism to maintain public trust.

cs.CL↗

arXiv2025

OWL: Probing Cross-Lingual Recall of Memorized Texts via World Literature

Large language models (LLMs) are known to memorize and recall English text from their pretraining data. However, the extent to which this ability generalizes to non-English languages or transfers across languages remains unclear. This paper investigates multilingual and cross-lingual memorization in LLMs, probing if memorized content in one language (e.g., English) can be recalled when presented in translation. To do so, we introduce OWL, a dataset of 31.5K aligned excerpts from 20 books in ten languages, including English originals, official translations (Vietnamese, Spanish, Turkish), and new translations in six low-resource languages (Sesotho, Yoruba, Maithili, Malagasy, Setswana, Tahitian). We evaluate memorization across model families and sizes through three tasks: (1) direct probing, which asks the model to identify a book's title and author; (2) name cloze, which requires predicting masked character names; and (3) prefix probing, which involves generating continuations. We find that LLMs consistently recall content across languages, even for texts without direct translation in pretraining data. GPT-4o, for example, identifies authors and titles 69% of the time and masked entities 6% of the time in newly translated excerpts. Perturbations (e.g., masking characters, shuffling words) modestly reduce direct probing accuracy (7% drop for shuffled official translations). Our results highlight the extent of cross-lingual memorization and provide insights on the differences between the models.

cs.CL↗

arXiv2025

One ruler to measure them all: Benchmarking multilingual long-context language models

We present ONERULER, a multilingual benchmark designed to evaluate long-context language models across 26 languages. ONERULER adapts the English-only RULER benchmark (Hsieh et al., 2024) by including seven synthetic tasks that test both retrieval and aggregation, including new variations of the "needle-in-a-haystack" task that allow for the possibility of a nonexistent needle. We create ONERULER through a two-step process, first writing English instructions for each task and then collaborating with native speakers to translate them into 25 additional languages. Experiments with both open-weight and closed LLMs reveal a widening performance gap between low- and high-resource languages as context length increases from 8K to 128K tokens. Surprisingly, English is not the top-performing language on long-context tasks (ranked 6th out of 26), with Polish emerging as the top language. Our experiments also show that many LLMs (particularly OpenAI's o3-mini-high) incorrectly predict the absence of an answer, even in high-resource languages. Finally, in cross-lingual scenarios where instructions and context appear in different languages, performance can fluctuate by up to 20% depending on the instruction language. We hope the release of ONERULER will facilitate future research into improving multilingual and cross-lingual long-context training pipelines.

cs.CL↗

arXiv2025

Does quantization affect models' performance on long-context tasks?

Large language models (LLMs) now support context windows exceeding 128K tokens, but this comes with significant memory requirements and high inference latency. Quantization can mitigate these costs, but may degrade performance. In this work, we present the first systematic evaluation of quantized LLMs on tasks with long inputs (>64K tokens) and long-form outputs. Our evaluation spans 9.7K test examples, five quantization methods (FP8, GPTQ-int8, AWQ-int4, GPTQ-int4, BNB-nf4), and five models (Llama-3.1 8B and 70B; Qwen-2.5 7B, 32B, and 72B). We find that, on average, 8-bit quantization preserves accuracy (~0.8% drop), whereas 4-bit methods lead to substantial losses, especially for tasks involving long-context inputs (drops of up to 59%). This degradation tends to worsen when the input is in a language other than English. Crucially, the effects of quantization depend heavily on the quantization method, model, and task. For instance, while Qwen-2.5 72B remains robust under BNB-nf4, Llama-3.1 70B experiences a 32% performance drop on the same task. These findings highlight the importance of a careful, task-specific evaluation before deploying quantized LLMs, particularly in long-context scenarios and for languages other than English.

cs.CL↗

arXiv2025

Preliminary Ranking of WMT25 General Machine Translation Systems

We present the preliminary rankings of machine translation (MT) systems submitted to the WMT25 General Machine Translation Shared Task, as determined by automatic evaluation metrics. Because these rankings are derived from automatic evaluation, they may exhibit a bias toward systems that employ re-ranking techniques, such as Quality Estimation or Minimum Bayes Risk decoding. The official WMT25 ranking will be based on human evaluation, which is more reliable and will supersede these results. The official WMT25 ranking will be based on human evaluation, which is more reliable and will supersede these results. The purpose of releasing these findings now is to assist task participants with their system description papers; not to provide final findings.

cs.CL↗

arXiv2025

An Interdisciplinary Approach to Human-Centered Machine Translation

Machine Translation (MT) tools are widely used today, often in contexts where professional translators are not present. Despite progress in MT technology, a gap persists between system development and real-world usage, particularly for non-expert users who may struggle to assess translation reliability. This paper advocates for a human-centered approach to MT, emphasizing the alignment of system design with diverse communicative goals and contexts of use. We survey the literature in Translation Studies and Human-Computer Interaction to recontextualize MT evaluation and design to address the diverse real-world scenarios in which MT is used today.

cs.CL↗

arXiv2025

CaLMQA: Exploring culturally specific long-form question answering across 23 languages

Despite rising global usage of large language models (LLMs), their ability to generate long-form answers to culturally specific questions remains unexplored in many languages. To fill this gap, we perform the first study of textual multilingual long-form QA by creating CaLMQA, a dataset of 51.7K culturally specific questions across 23 different languages. We define culturally specific questions as those that refer to concepts unique to one or a few cultures, or have different answers depending on the cultural or regional context. We obtain these questions by crawling naturally-occurring questions from community web forums in high-resource languages, and by hiring native speakers to write questions in under-resourced, rarely-studied languages such as Fijian and Kirundi. Our data collection methodologies are translation-free, enabling the collection of culturally unique questions like "Kuber iki umwami wa mbere w'uburundi yitwa Ntare?" (Kirundi; English translation: "Why was the first king of Burundi called Ntare (Lion)?"). We evaluate factuality, relevance and surface-level quality of LLM-generated long-form answers, finding that (1) for many languages, even the best models make critical surface-level errors (e.g., answering in the wrong language, repetition), especially for low-resource languages; and (2) answers to culturally specific questions contain more factual errors than answers to culturally agnostic questions -- questions that have consistent meaning and answer across many cultures. We release CaLMQA to facilitate future research in cultural and multilingual long-form QA.

cs.CL↗

arXiv2025

People who frequently use ChatGPT for writing tasks are accurate and robust detectors of AI-generated text

In this paper, we study how well humans can detect text generated by commercial LLMs (GPT-4o, Claude, o1). We hire annotators to read 300 non-fiction English articles, label them as either human-written or AI-generated, and provide paragraph-length explanations for their decisions. Our experiments show that annotators who frequently use LLMs for writing tasks excel at detecting AI-generated text, even without any specialized training or feedback. In fact, the majority vote among five such "expert" annotators misclassifies only 1 of 300 articles, significantly outperforming most commercial and open-source detectors we evaluated even in the presence of evasion tactics like paraphrasing and humanization. Qualitative analysis of the experts' free-form explanations shows that while they rely heavily on specific lexical clues ('AI vocabulary'), they also pick up on more complex phenomena within the text (e.g., formality, originality, clarity) that are challenging to assess for automatic detectors. We release our annotated dataset and code to spur future research into both human and automated detection of AI-generated text.

cs.CL↗

arXiv2024

Aurora-M: Open Source Continual Pre-training for Multilingual Language and Code

Pretrained language models are an integral part of AI applications, but their high computational cost for training limits accessibility. Initiatives such as Bloom and StarCoder aim to democratize access to pretrained models for collaborative community development. Despite these efforts, such models encounter challenges such as limited multilingual capabilities, risks of catastrophic forgetting during continual pretraining, and the high costs of training models from scratch, alongside the need to align with AI safety standards and regulatory frameworks. This paper presents Aurora-M, a 15B parameter multilingual open-source model trained on English, Finnish, Hindi, Japanese, Vietnamese, and code. Continually pretrained from StarCoderPlus on 435B additional tokens, Aurora-M surpasses 2T tokens in total training token count. It is the first open-source multilingual model fine-tuned on human-reviewed safety instructions, thus aligning its development not only with conventional red-teaming considerations, but also with the specific concerns articulated in the Biden-Harris Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence. We evaluate Aurora-M across a wide range of tasks and languages, showcasing its robustness against catastrophic forgetting and its superior performance in multilingual settings, particularly in safety evaluations. We open-source Aurora-M and its variants to encourage responsible open-source development of large language models at https://huggingface.co/aurora-m.

cs.CL↗

arXiv2024

One Thousand and One Pairs: A "novel" challenge for long-context language models

Synthetic long-context LLM benchmarks (e.g., "needle-in-the-haystack") test only surface-level retrieval capabilities, but how well can long-context LLMs retrieve, synthesize, and reason over information across book-length inputs? We address this question by creating NoCha, a dataset of 1,001 minimally different pairs of true and false claims about 67 recently-published English fictional books, written by human readers of those books. In contrast to existing long-context benchmarks, our annotators confirm that the largest share of pairs in NoCha require global reasoning over the entire book to verify. Our experiments show that while human readers easily perform this task, it is enormously challenging for all ten long-context LLMs that we evaluate: no open-weight model performs above random chance (despite their strong performance on synthetic benchmarks), while GPT-4o achieves the highest accuracy at 55.8%. Further analysis reveals that (1) on average, models perform much better on pairs that require only sentence-level retrieval vs. global reasoning; (2) model-generated explanations for their decisions are often inaccurate even for correctly-labeled claims; and (3) models perform substantially worse on speculative fiction books that contain extensive world-building. The methodology proposed in NoCha allows for the evolution of the benchmark dataset and the easy analysis of future models.

cs.CL↗

arXiv2024

Error Span Annotation: A Balanced Approach for Human Evaluation of Machine Translation

High-quality Machine Translation (MT) evaluation relies heavily on human judgments. Comprehensive error classification methods, such as Multidimensional Quality Metrics (MQM), are expensive as they are time-consuming and can only be done by experts, whose availability may be limited especially for low-resource languages. On the other hand, just assigning overall scores, like Direct Assessment (DA), is simpler and faster and can be done by translators of any level, but is less reliable. In this paper, we introduce Error Span Annotation (ESA), a human evaluation protocol which combines the continuous rating of DA with the high-level error severity span marking of MQM. We validate ESA by comparing it to MQM and DA for 12 MT systems and one human reference translation (English to German) from WMT23. The results show that ESA offers faster and cheaper annotations than MQM at the same quality level, without the requirement of expensive MQM experts.

cs.CL↗

arXiv2024

FABLES: Evaluating faithfulness and content selection in book-length summarization

While long-context large language models (LLMs) can technically summarize book-length documents (>100K tokens), the length and complexity of the documents have so far prohibited evaluations of input-dependent aspects like faithfulness. In this paper, we conduct the first large-scale human evaluation of faithfulness and content selection on LLM-generated summaries of fictional books. Our study mitigates the issue of data contamination by focusing on summaries of books published in 2023 or 2024, and we hire annotators who have fully read each book prior to the annotation task to minimize cost and cognitive burden. We collect FABLES, a dataset of annotations on 3,158 claims made in LLM-generated summaries of 26 books, at a cost of $5.2K USD, which allows us to rank LLM summarizers based on faithfulness: Claude-3-Opus significantly outperforms all closed-source LLMs, while the open-source Mixtral is on par with GPT-3.5-Turbo. An analysis of the annotations reveals that most unfaithful claims relate to events and character states, and they generally require indirect reasoning over the narrative to invalidate. While LLM-based auto-raters have proven reliable for factuality and coherence in other settings, we implement several LLM raters of faithfulness and find that none correlates strongly with human annotations, especially with regard to detecting unfaithful claims. Our experiments suggest that detecting unfaithful claims is an important future direction not only for summarization evaluation but also as a testbed for long-context understanding. Finally, we move beyond faithfulness by exploring content selection errors in book-length summarization: we develop a typology of omission errors related to crucial narrative elements and also identify a systematic over-emphasis on events occurring towards the end of the book.

cs.CL↗

arXiv2024

Preliminary WMT24 Ranking of General MT Systems and LLMs

This is the preliminary ranking of WMT24 General MT systems based on automatic metrics. The official ranking will be a human evaluation, which is superior to the automatic ranking and supersedes it. The purpose of this report is not to interpret any findings but only provide preliminary results to the participants of the General MT task that may be useful during the writing of the system submission.

cs.CL↗

arXiv2023

Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defense

The rise in malicious usage of large language models, such as fake content creation and academic plagiarism, has motivated the development of approaches that identify AI-generated text, including those based on watermarking or outlier detection. However, the robustness of these detection algorithms to paraphrases of AI-generated text remains unclear. To stress test these detectors, we build a 11B parameter paraphrase generation model (DIPPER) that can paraphrase paragraphs, condition on surrounding context, and control lexical diversity and content reordering. Using DIPPER to paraphrase text generated by three large language models (including GPT3.5-davinci-003) successfully evades several detectors, including watermarking, GPTZero, DetectGPT, and OpenAI's text classifier. For example, DIPPER drops detection accuracy of DetectGPT from 70.3% to 4.6% (at a constant false positive rate of 1%), without appreciably modifying the input semantics. To increase the robustness of AI-generated text detection to paraphrase attacks, we introduce a simple defense that relies on retrieving semantically-similar generations and must be maintained by a language model API provider. Given a candidate text, our algorithm searches a database of sequences previously generated by the API, looking for sequences that match the candidate text within a certain threshold. We empirically verify our defense using a database of 15M generations from a fine-tuned T5-XXL model and find that it can detect 80% to 97% of paraphrased generations across different settings while only classifying 1% of human-written sequences as AI-generated. We open-source our models, code and data.

cs.CL↗

arXiv2023

Large language models effectively leverage document-level context for literary translation, but critical errors persist

Large language models (LLMs) are competitive with the state of the art on a wide range of sentence-level translation datasets. However, their ability to translate paragraphs and documents remains unexplored because evaluation in these settings is costly and difficult. We show through a rigorous human evaluation that asking the Gpt-3.5 (text-davinci-003) LLM to translate an entire literary paragraph (e.g., from a novel) at once results in higher-quality translations than standard sentence-by-sentence translation across 18 linguistically-diverse language pairs (e.g., translating into and out of Japanese, Polish, and English). Our evaluation, which took approximately 350 hours of effort for annotation and analysis, is conducted by hiring translators fluent in both the source and target language and asking them to provide both span-level error annotations as well as preference judgments of which system's translations are better. We observe that discourse-level LLM translators commit fewer mistranslations, grammar errors, and stylistic inconsistencies than sentence-level approaches. With that said, critical errors still abound, including occasional content omissions, and a human translator's intervention remains necessary to ensure that the author's voice remains intact. We publicly release our dataset and error annotations to spur future research on evaluation of document-level literary translation.

cs.CL↗

arXiv2022

NarrativeTime: Dense Temporal Annotation on a Timeline

For the past decade, temporal annotation has been sparse: only a small portion of event pairs in a text was annotated. We present NarrativeTime, the first timeline-based annotation framework that achieves full coverage of all possible TLinks. To compare with the previous SOTA in dense temporal annotation, we perform full re-annotation of TimeBankDense corpus, which shows comparable agreement with a significant increase in density. We contribute TimeBankNT corpus (with each text fully annotated by two expert annotators), extensive annotation guidelines, open-source tools for annotation and conversion to TimeML format, baseline results, as well as quantitative and qualitative analysis of inter-annotator agreement.

cs.CL↗