BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Hinrich Schütze

Hinrich Schütze contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Computation and LanguageArtificial IntelligenceMachine LearningMultiagent Systems

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
43works
0followers
4topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

43 published item(s)

preprint2026arXiv

Calibration Is Not Enough: Evaluating Confidence Estimation Under Language Variations

Confidence estimation (CE) indicates how reliable the answers of large language models (LLMs) are, and can impact user trust and decision-making. Existing work evaluates CE methods almost exclusively through calibration, examining whether stated confidence aligns with accuracy, or discrimination, whether confidence is ranked higher for correct predictions than incorrect ones. However, these facets ignore pitfalls of CE in the context of LLMs and language variation: confidence estimates should remain consistent under semantically equivalent prompt or answer variations, and should change when the answer meaning differs. Therefore, we present a comprehensive evaluation framework for CE that measures their confidence quality on three new aspects: robustness of confidence against prompt perturbations, stability across semantic equivalent answers, and sensitivity to semantically different answers. In our work, we demonstrate that common CE methods for LLMs often fail on these metrics: methods that achieve good performance on calibration or discrimination are not robust to prompt variations or are not sensitive to answer changes. Overall, our framework reveals limitations of existing CE evaluations relevant for real-world LLM use cases and provides practical guidance for selecting and designing more reliable CE methods.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Crosslingual On-Policy Self-Distillation for Multilingual Reasoning

Large language models (LLMs) have achieved remarkable progress in mathematical reasoning, but this ability is not equally accessible across languages. Especially low-resource languages exhibit much lower reasoning performance. To address this, we propose Crosslingual On-Policy Self-Distillation (COPSD), which transfers a model's own high-resource reasoning behavior to low-resource languages. COPSD uses the same model as student and teacher: the student sees only the low-resource problem, while the teacher receives privileged crosslingual context, including the problem translation and reference solution in English. Training minimizes full-distribution token-level divergence on the student's own rollouts, providing dense supervision while avoiding the sparsity and instability of outcome-only reinforcement learning (RL). Experiments on 17 low-resource African languages show that COPSD consistently improves low-resource mathematical reasoning across model sizes and substantially outperforms Group Relative Policy Optimization (GRPO). Further analyses show that COPSD improves answer-format adherence, strengthens test-time scaling, and generalizes to harder multilingual reasoning benchmarks, with especially large gains for lower-resource languages. We make our code and data available at: https://github.com/cisnlp/COPSD.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

DiM\textsuperscript{3}: Bridging Multilingual and Multimodal Models via Direction- and Magnitude-Aware Merging

Towards more general and human-like intelligence, large language models should seamlessly integrate both multilingual and multimodal capabilities; however, extending an existing multimodal model to many languages typically requires expensive multilingual multimodal data construction and repeated end-to-end retraining. We study a training-free alternative: injecting multilingual capability into an existing multimodal model by composing residual updates in the shared language model backbone. The key challenge is that multilingual and multimodal updates are heterogeneous, reflecting different functional roles in the shared model. To address this, we propose Direction- and Magnitude-aware Multilingual Multimodal merging (DiM3), which selectively composes the two updates at each parameter dimension while preserving the original vision encoder and multimodal projector. Experiments on multilingual benchmarks in both text-only and vision-language settings, covering 57 languages across LLaVA- and Qwen-based backbones, show that DiM3 consistently outperforms existing merging baselines, substantially improves multilingual performance over the original multimodal model, and remains competitive with dedicated multilingual multimodal fine-tuning while largely retaining general multimodal ability. We further show that DiM3 can be directly applied to already trained multilingual multimodal models and still yield additional gains. Further interpretability analysis shows that DiM3 primarily reshapes intermediate-layer semantic representations, strengthening cross-lingual alignment under both text-only and multimodal inputs while preserving higher-layer task-sensitive structure. Our repository is on https://github.com/wzj1718/DiM3.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Evaluating Contextually Mediated Factual Recall in Multilingual Large Language Models

Large language models (LLMs) can recall a wide range of factual knowledge across languages. However, existing factual recall evaluations primarily assess fact retrieval in isolation, where the queried entity is explicitly named and the fact is requested directly. In natural language use, facts are often accessed through context, where the relevant entity is introduced only indirectly. In this work, we study contextually mediated factual recall, asking whether LLMs can reliably retrieve factual knowledge when the target entity is embedded in a naturalistic context rather than queried explicitly, across languages. We construct controlled prompts that preserve the underlying fact while introducing referential mediation through contextual sentences. To disentangle contextual effects from name-specific associations, we further compare performance using synthetic names and real names across languages. Evaluating multiple model families in five languages, we find that contextual mediation consistently degrades factual recall, with substantial variation across relations. Larger models are more robust to contextual mediation, exhibiting a reduced performance gap relative to direct queries, while the effect of real names and name origin is mixed and unsystematic. These findings highlight a gap between isolated factual recall and context-dependent language understanding in multilingual LLMs.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

GKnow: Measuring the Entanglement of Gender Bias and Factual Gender

Recent works have analyzed the impact of individual components of neural networks on gendered predictions, often with a focus on mitigating gender bias. However, mechanistic interpretations of gender tend to (i) focus on a very specific gender-related task, such as gendered pronoun prediction, or (ii) fail to distinguish between the production of factually gendered outputs (the correct assumption of gender given a word that carries gender as a semantic property) and gender biased outputs (based on a stereotype). To address these issues, we curate \gknow, a benchmark to assess gender knowledge and gender bias in language models across different types of gender-related predictions. \gknow allows us to identify and analyze circuits and individual neurons responsible for gendered predictions. We test the impact of neuron ablation on benchmarks for disentangling stereotypical and factual gender (DiFair and the test set of GKnow), as well as StereoSet. Results show that gender bias and factual gender are severely entangled on the level of both circuits and neurons, entailing that ablation is an unreliable debiasing method. Furthermore, we show that benchmarks for evaluating gender bias can hide the decrease in factual gender knowledge that accompanies neuron ablation. We curate GKnow as a contribution to the continuous development of robust gender bias benchmarks.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

High-Rank Structured Modulation for Parameter-Efficient Fine-Tuning

As the number of model parameters increases, parameter-efficient fine-tuning (PEFT) has become the go-to choice for tailoring pre-trained large language models. Low-rank Adaptation (LoRA) uses a low-rank update method to simulate full parameter fine-tuning, which is widely used to reduce resource requirements. However, decreasing the rank encounters challenges with limited representational capacity when compared to full parameter fine-tuning. We present \textbf{SMoA}, a high-rank \textbf{S}tructured \textbf{MO}dulation \textbf{A}dapter that uses fewer trainable parameters while maintaining a higher rank, thereby improving the model's representational capacity and offering improved performance potential. The core idea is to freeze the original pretrained weights and selectively amplify or suppress important features of the original weights across multiple subspaces. The subspace mechanism provides an efficient way to increase the capacity and complexity of a model. We conduct both theoretical analyses and empirical studies on various tasks. Experiment results show that SMoA outperforms LoRA and its variants on 10 tasks, with extensive ablation studies validating its effectiveness.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Left, Right, or Center? Evaluating LLM Framing in News Classification and Generation

Large Language Model (LLM) based summarization and text generation are increasingly used for producing and rewriting text, raising concerns about political framing in journalism where subtle wording choices can shape interpretation. Across nine state-of-the-art LLMs, we study political framing by testing whether LLMs' classification-based bias signals align with framing behavior in their generated summaries. We first compare few-shot ideology predictions against LEFT/CENTER/RIGHT labels. We then generate "steered" summaries under FAITHFUL, CENTRIST, LEFT, and RIGHT prompts, and score all outputs using a single fixed ideology evaluator. We find pervasive ideological center-collapse in both article-level ratings and generated text, indicating a systematic tendency toward centrist framing. Among evaluated models, Grok 4 is by far the most ideologically expressive generator, while Claude Sonnet 4.5 and Llama 3.1 achieve the strongest bias-rating performance among commercial and open-weight models, respectively.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning

Large Language Models (LLMs) have demonstrated impressive capabilities across a wide range of NLP tasks, but they remain fundamentally stateless, constrained by limited context windows that hinder long-horizon reasoning. Recent efforts to address this limitation often augment LLMs with an external memory bank, yet most existing pipelines are static and heuristic-driven, lacking a learned mechanism for deciding what to store, update, or retrieve. We present Memory-R1, a reinforcement learning (RL) framework that equips LLMs with the ability to actively manage and utilize external memory through two specialized agents: a Memory Manager that learns structured operations, including ADD, UPDATE, DELETE, and NOOP; and an Answer Agent that pre-selects and reasons over relevant entries. Both agents are fine-tuned with outcome-driven RL (PPO and GRPO), enabling adaptive memory management with minimal supervision. With only 152 training QA pairs, Memory-R1 outperforms strong baselines and generalizes across diverse question types, three benchmarks (LoCoMo, MSC, LongMemEval), and multiple model scales (3B-14B).

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

PlaM: Training-Free Plateau-Guided Model Merging for Better Visual Grounding in MLLMs

Multimodal Large Language Models (MLLMs) rely on strong linguistic reasoning inherited from their base language models. However, multimodal instruction fine-tuning paradoxically degrades this text's reasoning capability, undermining multimodal performance. To address this issue, we propose a training-free framework to mitigate this degradation. Through layer-wise vision token masking, we reveal a common three-stage pattern in multimodal large language models: early-modal separation, mid-modal alignment, and late-modal degradation. By analyzing the behavior of MLLMs at different stages, we propose a plateau-guided model merging method that selectively injects base language model parameters into MLLMs. Experimental results based on five MLLMs on nine benchmarks demonstrate the effectiveness of our method. Attention-based analysis further reveals that merging shifts attention from diffuse, scattered patterns to focused localization on task-relevant visual regions. Our repository is on https://github.com/wzj1718/PlaM.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Relational Linearity is a Predictor of Hallucinations

Hallucination is a central failure mode in large language models (LLMs). We focus on hallucinations of answers to questions like: "Which instrument did Glenn Gould play?", but we ask these questions for synthetic entities that are unknown to the model. Surprisingly, we find that medium-size models like Gemma-7B-IT frequently hallucinate, i.e., they have difficulty recognizing that the hallucinated fact is not part of their knowledge. We hypothesize that an important factor in causing these hallucinations is the linearity of the relation: linear relations tend to be stored more abstractly, making it difficult for the LLM to assess its knowledge; the facts of nonlinear relations tend to be stored more directly, making knowledge assessment easier. To investigate this hypothesis, we create SyntHal, a dataset of 6000 synthetic entities for six relations. In our experiments with four models, we determine, for each relation, the hallucination rate on SyntHal and also measure its linearity, using $Δ\cos$. We find a strong correlation ($r \in [.78,.82]$) between relational linearity and hallucination rate, providing evidence for our hypothesis that the underlying storage of triples of a relation is a factor in how well a model can self-assess its knowledge. This finding has implications for how to manage hallucination behavior and suggests new research directions for improving the representation of factual knowledge in LLMs.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

SAD: A Large-Scale Strategic Argumentative Dialogue Dataset

Argumentation generation has attracted substantial research interest due to its central role in human reasoning and decision-making. However, most existing argumentative corpora focus on non-interactive, single-turn settings, either generating arguments from a given topic or refuting an existing argument. In practice, however, argumentation is often realized as multi-turn dialogue, where speakers defend their stances and employ diverse argumentative strategies to strengthen persuasiveness. To support deeper modeling of argumentation dialogue, we present the first large-scale \textbf{S}trategic \textbf{A}rgumentative \textbf{D}ialogue dataset, SAD, consisting of 392,822 examples. Grounded in argumentation theories, we annotate each utterance with five strategy types, allowing multiple strategies per utterance. Unlike prior datasets, SAD requires models to generate contextually appropriate arguments conditioned on the dialogue history, a specified stance on the topic, and targeted argumentation strategies. We further benchmark a range of pretrained generative models on SAD and present in-depth analysis of strategy usage patterns in argumentation.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Through the LLM Looking Glass: A Socratic Probing of Donkeys, Elephants, and Markets

Large Language Models (LLMs) are widely used for text generation, making it crucial to address potential bias. This study investigates ideological framing bias in LLM-generated articles, focusing on the subtle and subjective nature of such bias in journalistic contexts. We evaluate eight widely used LLMs on two datasets-POLIGEN and ECONOLEX-covering political and economic discourse where framing bias is most pronounced. Beyond text generation, LLMs are increasingly used as evaluators (LLM-as-a-judge), providing feedback that can shape human judgment or inform newer model versions. Inspired by the Socratic method, we further analyze LLMs' feedback on their own outputs to identify inconsistencies in their reasoning. Our results show that most LLMs can accurately annotate ideologically framed text, with GPT-4o achieving human-level accuracy and high agreement with human annotators. However, Socratic probing reveals that when confronted with binary comparisons, LLMs often exhibit preference toward one perspective or perceive certain viewpoints as less biased.

0 citations0 reviewsNo code linkedOpen access
preprint2023arXiv

NLNDE at SemEval-2023 Task 12: Adaptive Pretraining and Source Language Selection for Low-Resource Multilingual Sentiment Analysis

This paper describes our system developed for the SemEval-2023 Task 12 "Sentiment Analysis for Low-resource African Languages using Twitter Dataset". Sentiment analysis is one of the most widely studied applications in natural language processing. However, most prior work still focuses on a small number of high-resource languages. Building reliable sentiment analysis systems for low-resource languages remains challenging, due to the limited training data in this task. In this work, we propose to leverage language-adaptive and task-adaptive pretraining on African texts and study transfer learning with source language selection on top of an African language-centric pretrained language model. Our key findings are: (1) Adapting the pretrained model to the target language and task using a small yet relevant corpus improves performance remarkably by more than 10 F1 score points. (2) Selecting source languages with positive transfer gains during training can avoid harmful interference from dissimilar languages, leading to better results in multilingual and cross-lingual settings. In the shared task, our system wins 8 out of 15 tracks and, in particular, performs best in the multilingual evaluation.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Analyzing Hate Speech Data along Racial, Gender and Intersectional Axes

To tackle the rising phenomenon of hate speech, efforts have been made towards data curation and analysis. When it comes to analysis of bias, previous work has focused predominantly on race. In our work, we further investigate bias in hate speech datasets along racial, gender and intersectional axes. We identify strong bias against African American English (AAE), masculine and AAE+Masculine tweets, which are annotated as disproportionately more hateful and offensive than from other demographics. We provide evidence that BERT-based models propagate this bias and show that balancing the training data for these protected attributes can lead to fairer models with regards to gender, but not race.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

CaMEL: Case Marker Extraction without Labels

We introduce CaMEL (Case Marker Extraction without Labels), a novel and challenging task in computational morphology that is especially relevant for low-resource languages. We propose a first model for CaMEL that uses a massively multilingual corpus to extract case markers in 83 languages based only on a noun phrase chunker and an alignment system. To evaluate CaMEL, we automatically construct a silver standard from UniMorph. The case markers extracted by our model can be used to detect and visualise similarities and differences between the case systems of different languages as well as to annotate fine-grained deep cases in languages in which they are not overtly marked.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

CoDA21: Evaluating Language Understanding Capabilities of NLP Models With Context-Definition Alignment

Pretrained language models (PLMs) have achieved superhuman performance on many benchmarks, creating a need for harder tasks. We introduce CoDA21 (Context Definition Alignment), a challenging benchmark that measures natural language understanding (NLU) capabilities of PLMs: Given a definition and a context each for k words, but not the words themselves, the task is to align the k definitions with the k contexts. CoDA21 requires a deep understanding of contexts and definitions, including complex inference and world knowledge. We find that there is a large gap between human and PLM performance, suggesting that CoDA21 measures an aspect of NLU that is not sufficiently covered in existing benchmarks.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Domain Adaptation for Sparse-Data Settings: What Do We Gain by Not Using Bert?

The practical success of much of NLP depends on the availability of training data. However, in real-world scenarios, training data is often scarce, not least because many application domains are restricted and specific. In this work, we compare different methods to handle this problem and provide guidelines for building NLP applications when there is only a small amount of labeled training data available for a specific domain. While transfer learning with pre-trained language models outperforms other methods across tasks, alternatives do not perform much worse while requiring much less computational effort, thus significantly reducing monetary and environmental cost. We examine the performance tradeoffs of several such alternatives, including models that can be trained up to 175K times faster and do not require a single GPU.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Don't Forget Cheap Training Signals Before Building Unsupervised Bilingual Word Embeddings

Bilingual Word Embeddings (BWEs) are one of the cornerstones of cross-lingual transfer of NLP models. They can be built using only monolingual corpora without supervision leading to numerous works focusing on unsupervised BWEs. However, most of the current approaches to build unsupervised BWEs do not compare their results with methods based on easy-to-access cross-lingual signals. In this paper, we argue that such signals should always be considered when developing unsupervised BWE methods. The two approaches we find most effective are: 1) using identical words as seed lexicons (which unsupervised approaches incorrectly assume are not available for orthographically distinct language pairs) and 2) combining such lexicons with pairs extracted by matching romanized versions of words with an edit distance threshold. We experiment on thirteen non-Latin languages (and English) and show that such cheap signals work well and that they outperform using more complex unsupervised methods on distant language pairs such as Chinese, Japanese, Kannada, Tamil, and Thai. In addition, they are even competitive with the use of high-quality lexicons in supervised approaches. Our results show that these training signals should not be neglected when building BWEs, even for distant languages.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Flow-Adapter Architecture for Unsupervised Machine Translation

In this work, we propose a flow-adapter architecture for unsupervised NMT. It leverages normalizing flows to explicitly model the distributions of sentence-level latent representations, which are subsequently used in conjunction with the attention mechanism for the translation task. The primary novelties of our model are: (a) capturing language-specific sentence representations separately for each language using normalizing flows and (b) using a simple transformation of these latent representations for translating from one language to another. This architecture allows for unsupervised training of each language independently. While there is prior work on latent variables for supervised MT, to the best of our knowledge, this is the first work that uses latent variables and normalizing flows for unsupervised MT. We obtain competitive results on several unsupervised MT benchmarks.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Graph Neural Networks for Multiparallel Word Alignment

After a period of decrease, interest in word alignments is increasing again for their usefulness in domains such as typological research, cross-lingual annotation projection, and machine translation. Generally, alignment algorithms only use bitext and do not make use of the fact that many parallel corpora are multiparallel. Here, we compute high-quality word alignments between multiple language pairs by considering all language pairs together. First, we create a multiparallel word alignment graph, joining all bilingual word alignment pairs in one graph. Next, we use graph neural networks (GNNs) to exploit the graph structure. Our GNN approach (i) utilizes information about the meaning, position, and language of the input words, (ii) incorporates information from multiple parallel sentences, (iii) adds and removes edges from the initial alignments, and (iv) yields a prediction model that can generalize beyond the training sentences. We show that community detection provides valuable information for multiparallel word alignment. Our method outperforms previous work on three word-alignment datasets and on a downstream task.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

LMTurk: Few-Shot Learners as Crowdsourcing Workers in a Language-Model-as-a-Service Framework

Vast efforts have been devoted to creating high-performance few-shot learners, i.e., large-scale pretrained language models (PLMs) that perform well with little downstream task training data. Training PLMs has incurred significant cost, but utilizing the few-shot learners is still challenging due to their enormous size. This work focuses on a crucial question: How to make effective use of these few-shot learners? We propose LMTurk, a novel approach that treats few-shot learners as crowdsourcing workers. The rationale is that crowdsourcing workers are in fact few-shot learners: They are shown a few illustrative examples to learn about a task and then start annotating. LMTurk employs few-shot learners built upon PLMs as workers. We show that the resulting annotations can be utilized to train models that solve the task well and are small enough to be deployable in practical scenarios. Active learning is integrated into LMTurk to reduce the amount of queries made to PLMs, minimizing the computational cost of running PLM inference passes. Altogether, LMTurk is an important step towards making effective use of current PLMs.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Modular and Parameter-Efficient Multimodal Fusion with Prompting

Recent research has made impressive progress in large-scale multimodal pre-training. In the context of the rapid growth of model size, it is necessary to seek efficient and flexible methods other than finetuning. In this paper, we propose to use prompt vectors to align the modalities. Our method achieves comparable performance to several other multimodal fusion methods in low-resource settings. We further show that our method is modular and parameter-efficient for processing tasks involving two or more data modalities.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Semantic-Oriented Unlabeled Priming for Large-Scale Language Models

Due to the high costs associated with finetuning large language models, various recent works propose to adapt them to specific tasks without any parameter updates through in-context learning. Unfortunately, for in-context learning there is currently no way to leverage unlabeled data, which is often much easier to obtain in large quantities than labeled examples. In this work, we therefore investigate ways to make use of unlabeled examples to improve the zero-shot performance of pretrained language models without any finetuning: We introduce Semantic-Oriented Unlabeled Priming (SOUP), a method that classifies examples by retrieving semantically similar unlabeled examples, assigning labels to them in a zero-shot fashion, and then using them for in-context learning. We also propose bag-of-contexts priming, a new priming strategy that is more suitable for our setting and enables the usage of more examples than fit into the context window.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Towards a Broad Coverage Named Entity Resource: A Data-Efficient Approach for Many Diverse Languages

Parallel corpora are ideal for extracting a multilingual named entity (MNE) resource, i.e., a dataset of names translated into multiple languages. Prior work on extracting MNE datasets from parallel corpora required resources such as large monolingual corpora or word aligners that are unavailable or perform poorly for underresourced languages. We present CLC-BN, a new method for creating an MNE resource, and apply it to the Parallel Bible Corpus, a corpus of more than 1000 languages. CLC-BN learns a neural transliteration model from parallel-corpus statistics, without requiring any other bilingual resources, word aligners, or seed data. Experimental results show that CLC-BN clearly outperforms prior work. We release an MNE resource for 1340 languages and demonstrate its effectiveness in two downstream tasks: knowledge graph augmentation and bilingual lexicon induction.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Does He Wink or Does He Nod? A Challenging Benchmark for Evaluating Word Understanding of Language Models

Recent progress in pretraining language models on large corpora has resulted in large performance gains on many NLP tasks. These large models acquire linguistic knowledge during pretraining, which helps to improve performance on downstream tasks via fine-tuning. To assess what kind of knowledge is acquired, language models are commonly probed by querying them with `fill in the blank' style cloze questions. Existing probing datasets mainly focus on knowledge about relations between words and entities. We introduce WDLMPro (Word Definition Language Model Probing) to evaluate word understanding directly using dictionary definitions of words. In our experiments, three popular pretrained language models struggle to match words and their definitions. This indicates that they understand many words poorly and that our new probing task is a difficult challenge that could help guide research on LMs in the future.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference

Some NLP tasks can be solved in a fully unsupervised fashion by providing a pretrained language model with "task descriptions" in natural language (e.g., Radford et al., 2019). While this approach underperforms its supervised counterpart, we show in this work that the two ideas can be combined: We introduce Pattern-Exploiting Training (PET), a semi-supervised training procedure that reformulates input examples as cloze-style phrases to help language models understand a given task. These phrases are then used to assign soft labels to a large set of unlabeled examples. Finally, standard supervised training is performed on the resulting training set. For several tasks and languages, PET outperforms supervised training and strong semi-supervised approaches in low-resource settings by a large margin.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Identifying Necessary Elements for BERT's Multilinguality

It has been shown that multilingual BERT (mBERT) yields high quality multilingual representations and enables effective zero-shot transfer. This is surprising given that mBERT does not use any crosslingual signal during training. While recent literature has studied this phenomenon, the reasons for the multilinguality are still somewhat obscure. We aim to identify architectural properties of BERT and linguistic properties of languages that are necessary for BERT to become multilingual. To allow for fast experimentation we propose an efficient setup with small BERT models trained on a mix of synthetic and natural data. Overall, we identify four architectural and two linguistic elements that influence multilinguality. Based on our insights, we experiment with a multilingual pretraining setup that modifies the masking strategy using VecMap, i.e., unsupervised embedding alignment. Experiments on XNLI with three languages indicate that our findings transfer from our small setup to larger scale settings.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Morphological Segmentation Inside-Out

Morphological segmentation has traditionally been modeled with non-hierarchical models, which yield flat segmentations as output. In many cases, however, proper morphological analysis requires hierarchical structure -- especially in the case of derivational morphology. In this work, we introduce a discriminative, joint model of morphological segmentation along with the orthographic changes that occur during word formation. To the best of our knowledge, this is the first attempt to approach discriminative segmentation with a context-free model. Additionally, we release an annotated treebank of 7454 English words with constituency parses, encouraging future research in this area.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Multilingual LAMA: Investigating Knowledge in Multilingual Pretrained Language Models

Recently, it has been found that monolingual English language models can be used as knowledge bases. Instead of structural knowledge base queries, masked sentences such as "Paris is the capital of [MASK]" are used as probes. We translate the established benchmarks TREx and GoogleRE into 53 languages. Working with mBERT, we investigate three questions. (i) Can mBERT be used as a multilingual knowledge base? Most prior work only considers English. Extending research to multiple languages is important for diversity and accessibility. (ii) Is mBERT's performance as knowledge base language-independent or does it vary from language to language? (iii) A multilingual model is trained on more text, e.g., mBERT is trained on 104 Wikipedias. Can mBERT leverage this for better performance? We find that using mBERT as a knowledge base yields varying performance across languages and pooling predictions across languages improves performance. Conversely, mBERT exhibits a language bias; e.g., when queried in Italian, it tends to predict Italy as the country of origin.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

A Hierarchical Decoder with Three-level Hierarchical Attention to Generate Abstractive Summaries of Interleaved Texts

Interleaved texts, where posts belonging to different threads occur in one sequence, are a common occurrence, e.g., online chat conversations. To quickly obtain an overview of such texts, existing systems first disentangle the posts by threads and then extract summaries from those threads. The major issues with such systems are error propagation and non-fluent summary. To address those, we propose an end-to-end trainable hierarchical encoder-decoder system. We also introduce a novel hierarchical attention mechanism which combines three levels of information from an interleaved text, i.e, posts, phrases and words, and implicitly disentangles the threads. We evaluated the proposed system on multiple interleaved text datasets, and it out-performs a SOTA two-step system by 20-40%.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Automatic Domain Adaptation Outperforms Manual Domain Adaptation for Predicting Financial Outcomes

In this paper, we automatically create sentiment dictionaries for predicting financial outcomes. We compare three approaches: (I) manual adaptation of the domain-general dictionary H4N, (ii) automatic adaptation of H4N and (iii) a combination consisting of first manual, then automatic adaptation. In our experiments, we demonstrate that the automatically adapted sentiment dictionary outperforms the previous state of the art in predicting the financial outcomes excess return and volatility. In particular, automatic adaptation performs better than manual adaptation. In our analysis, we find that annotation based on an expert's a priori belief about a word's meaning can be incorrect -- annotation should be performed based on the word's contexts in the target domain instead.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

BERTRAM: Improved Word Embeddings Have Big Impact on Contextualized Model Performance

Pretraining deep language models has led to large performance gains in NLP. Despite this success, Schick and Schütze (2020) recently showed that these models struggle to understand rare words. For static word embeddings, this problem has been addressed by separately learning representations for rare words. In this work, we transfer this idea to pretrained language models: We introduce BERTRAM, a powerful architecture based on BERT that is capable of inferring high-quality embeddings for rare words that are suitable as input representations for deep language models. This is achieved by enabling the surface form and contexts of a word to interact with each other in a deep architecture. Integrating BERTRAM into BERT leads to large performance increases due to improved representations of rare and medium frequency words on both a rare word probing task and three downstream tasks.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

E-BERT: Efficient-Yet-Effective Entity Embeddings for BERT

We present a novel way of injecting factual knowledge about entities into the pretrained BERT model (Devlin et al., 2019): We align Wikipedia2Vec entity vectors (Yamada et al., 2016) with BERT's native wordpiece vector space and use the aligned entity vectors as if they were wordpiece vectors. The resulting entity-enhanced version of BERT (called E-BERT) is similar in spirit to ERNIE (Zhang et al., 2019) and KnowBert (Peters et al., 2019), but it requires no expensive further pretraining of the BERT encoder. We evaluate E-BERT on unsupervised question answering (QA), supervised relation classification (RC) and entity linking (EL). On all three tasks, E-BERT outperforms BERT and other baselines. We also show quantitatively that the original BERT model is overly reliant on the surface form of entity names (e.g., guessing that someone with an Italian-sounding name speaks Italian), and that E-BERT mitigates this problem.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Extending Machine Language Models toward Human-Level Language Understanding

Language is crucial for human intelligence, but what exactly is its role? We take language to be a part of a system for understanding and communicating about situations. The human ability to understand and communicate about situations emerges gradually from experience and depends on domain-general principles of biological neural networks: connection-based learning, distributed representation, and context-sensitive, mutual constraint satisfaction-based processing. Current artificial language processing systems rely on the same domain general principles, embodied in artificial neural networks. Indeed, recent progress in this field depends on \emph{query-based attention}, which extends the ability of these systems to exploit context and has contributed to remarkable breakthroughs. Nevertheless, most current models focus exclusively on language-internal tasks, limiting their ability to perform tasks that depend on understanding situations. These systems also lack memory for the contents of prior situations outside of a fixed contextual span. We describe the organization of the brain's distributed understanding system, which includes a fast learning system that addresses the memory problem. We sketch a framework for future models of understanding drawing equally on cognitive neuroscience and artificial intelligence and exploiting query-based attention. We highlight relevant current directions and consider further developments needed to fully capture human-level language understanding in a computational system.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Inexpensive Domain Adaptation of Pretrained Language Models: Case Studies on Biomedical NER and Covid-19 QA

Domain adaptation of Pretrained Language Models (PTLMs) is typically achieved by unsupervised pretraining on target-domain text. While successful, this approach is expensive in terms of hardware, runtime and CO_2 emissions. Here, we propose a cheaper alternative: We train Word2Vec on target-domain text and align the resulting word vectors with the wordpiece vectors of a general-domain PTLM. We evaluate on eight biomedical Named Entity Recognition (NER) tasks and compare against the recently proposed BioBERT model. We cover over 60% of the BioBERT-BERT F1 delta, at 5% of BioBERT's CO_2 footprint and 2% of its cloud compute cost. We also show how to quickly adapt an existing general-domain Question Answering (QA) model to an emerging domain: the Covid-19 pandemic.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Multilingual Embeddings Jointly Induced from Contexts and Concepts: Simple, Strong and Scalable

Word embeddings induced from local context are prevalent in NLP. A simple and effective context-based multilingual embedding learner is Levy et al. (2017)'s S-ID (sentence ID) method. Another line of work induces high-performing multilingual embeddings from concepts (Dufter et al., 2018). In this paper, we propose Co+Co, a simple and scalable method that combines context-based and concept-based learning. From a sentence aligned corpus, concepts are extracted via sampling; words are then associated with their concept ID and sentence ID in embedding learning. This is the first work that successfully combines context-based and concept-based embedding learning. We show that Co+Co performs well for two different application scenarios: the Parallel Bible Corpus (1000+ languages, low-resource) and EuroParl (12 languages, high-resource). Among methods applicable to both corpora, Co+Co performs best in our evaluation setup of six tasks.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Multipurpose Intelligent Process Automation via Conversational Assistant

Intelligent Process Automation (IPA) is an emerging technology with a primary goal to assist the knowledge worker by taking care of repetitive, routine and low-cognitive tasks. Conversational agents that can interact with users in a natural language are potential application for IPA systems. Such intelligent agents can assist the user by answering specific questions and executing routine tasks that are ordinarily performed in a natural language (i.e., customer support). In this work, we tackle a challenge of implementing an IPA conversational assistant in a real-world industrial setting with a lack of structured training data. Our proposed system brings two significant benefits: First, it reduces repetitive and time-consuming activities and, therefore, allows workers to focus on more intelligent processes. Second, by interacting with users, it augments the resources with structured and to some extent labeled training data. We showcase the usage of the latter by re-implementing several components of our system with Transfer Learning (TL) methods.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Fly

Building on Petroni et al. (2019), we propose two new probing tasks analyzing factual knowledge stored in Pretrained Language Models (PLMs). (1) Negation. We find that PLMs do not distinguish between negated ("Birds cannot [MASK]") and non-negated ("Birds can [MASK]") cloze questions. (2) Mispriming. Inspired by priming methods in human psychology, we add "misprimes" to cloze questions ("Talk? Birds can [MASK]"). We find that PLMs are easily distracted by misprimes. These results suggest that PLMs still have a long way to go to adequately learn human-like factual knowledge.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Neural Semi-Markov Conditional Random Fields for Robust Character-Based Part-of-Speech Tagging

Character-level models of tokens have been shown to be effective at dealing with within-token noise and out-of-vocabulary words. But these models still rely on correct token boundaries. In this paper, we propose a novel end-to-end character-level model and demonstrate its effectiveness in multilingual settings and when token boundaries are noisy. Our model is a semi-Markov conditional random field with neural networks for character and segment representation. It requires no tokenizer. The model matches state-of-the-art baselines for various languages and significantly outperforms them on a noisy English version of a part-of-speech tagging benchmark dataset. Our code and the noisy dataset are publicly available at http://cistern.cis.lmu.de/semiCRF.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Sentence Meta-Embeddings for Unsupervised Semantic Textual Similarity

We address the task of unsupervised Semantic Textual Similarity (STS) by ensembling diverse pre-trained sentence encoders into sentence meta-embeddings. We apply, extend and evaluate different meta-embedding methods from the word embedding literature at the sentence level, including dimensionality reduction (Yin and Schütze, 2016), generalized Canonical Correlation Analysis (Rastogi et al., 2015) and cross-view auto-encoders (Bollegala and Bao, 2018). Our sentence meta-embeddings set a new unsupervised State of The Art (SoTA) on the STS Benchmark and on the STS12-STS16 datasets, with gains of between 3.7% and 6.4% Pearson's r over single-source systems.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Unsupervised Embedding-based Detection of Lexical Semantic Changes

This paper describes EmbLexChange, a system introduced by the "Life-Language" team for SemEval-2020 Task 1, on unsupervised detection of lexical-semantic changes. EmbLexChange is defined as the divergence between the embedding based profiles of word w (calculated with respect to a set of reference words) in the source and the target domains (source and target domains can be simply two time frames t1 and t2). The underlying assumption is that the lexical-semantic change of word w would affect its co-occurring words and subsequently alters the neighborhoods in the embedding spaces. We show that using a resampling framework for the selection of reference words, we can reliably detect lexical-semantic changes in English, German, Swedish, and Latin. EmbLexChange achieved second place in the binary detection of semantic changes in the SemEval-2020.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Ultradense Word Embeddings by Orthogonal Transformation

Embeddings are generic representations that are useful for many NLP tasks. In this paper, we introduce DENSIFIER, a method that learns an orthogonal transformation of the embedding space that focuses the information relevant for a task in an ultradense subspace of a dimensionality that is smaller by a factor of 100 than the original space. We show that ultradense embeddings generated by DENSIFIER reach state of the art on a lexicon creation task in which words are annotated with three types of lexical information - sentiment, concreteness and frequency. On the SemEval2015 10B sentiment analysis task we show that no information is lost when the ultradense subspace is used, but training is an order of magnitude more efficient due to the compactness of the ultradense space.

0 citations0 reviewsNo code linkedOpen access
preprint2015arXiv

AutoExtend: Extending Word Embeddings to Embeddings for Synsets and Lexemes

We present \textit{AutoExtend}, a system to learn embeddings for synsets and lexemes. It is flexible in that it can take any word embeddings as input and does not need an additional training corpus. The synset/lexeme embeddings obtained live in the same vector space as the word embeddings. A sparse tensor formalization guarantees efficiency and parallelizability. We use WordNet as a lexical resource, but AutoExtend can be easily applied to other resources like Freebase. AutoExtend achieves state-of-the-art performance on word similarity and word sense disambiguation tasks.

0 citations0 reviewsNo code linkedOpen access