Papers by Robert Litschko
Decoupling the Effect of Chain-of-Thought Reasoning: A Human Label Variation Perspective (2026.findings-acl)
Copied to clipboard
| Challenge: | Reasoning-tuned large language models (LLMs) with long Chain-of-Thought excel at single-answer tasks, yet their ability to model Human Label Variation remains underexplored. |
| Approach: | They conduct systematic disentanglement experiments to isolate the effect of reasoning text from intrinsic model priors on distribution-based tasks. |
| Outcome: | The proposed model improves distributional alignment, but distributional ranking is governed by model priors. |
Boosting Zero-shot Cross-lingual Retrieval by Training on Artificially Code-Switched Data (2023.findings-acl)
Copied to clipboard
| Challenge: | Using zero-shot rankers, cross-lingual IR models are limited by their language coverage. |
| Approach: | They propose to train ranking models on artificially code-switched data instead of using a dictionary. |
| Outcome: | The proposed approach is robust towards the ratio of code-switched tokens and extends to unseen languages. |
“Seeing the Big through the Small”: Can LLMs Approximate Human Judgment Distributions on NLI from a Few Explanations? (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Human label variation arises when multiple human annotators provide different labels for valid reasons. |
| Approach: | They propose to use crowd workers to represent human judgment distributions or expert linguists to provide detailed explanations for their chosen labels. |
| Outcome: | The proposed model can approximate human judgment distributions using a small number of expert labels and explanations. |
Reason to Rote: Rethinking Memorization in Reasoning (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models readily memorize arbitrary training instances, such as label noise . however, such memorization does not affect generalizable reasoning abilities . |
| Approach: | They investigate how large language models memorize label noise and why it affects generalizability. |
| Outcome: | The proposed model performs well on reasoning tasks even when memorized labels are missing . the proposed model is able to generalize to correctly answer "87+19=106" |
To Know or Not To Know? Analyzing Self-Consistency of Large Language Models under Ambiguity (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models (LLMs) have remarkable performance in a variety of tasks due to factual knowledge accumulated during pre-training. |
| Approach: | They propose an evaluation protocol that disentangles knowing from applying knowledge and test state-of-the-art LLMs on 49 ambiguous entities. |
| Outcome: | The proposed evaluation protocol disentangles knowing from applying knowledge and tests state-of-the-art LLMs on 49 ambiguous entities. |
Parameter-Efficient Neural Reranking for Cross-Lingual and Multilingual Retrieval (2022.coling-1)
Copied to clipboard
| Challenge: | State-of-the-art neural rankers are notoriously data-hungry and rarely used in multilingual and cross-lingual retrieval settings. |
| Approach: | They propose to use Sparse Fine-Tuning Masks and Adapters to transfer rankers trained on English data to other languages and cross-lingual setups by means of multilingual encoders. |
| Outcome: | The proposed methods outperform standard zero-shot transfer with full MMT fine-tuning while being more modular and reducing training times. |
Establishing Trustworthiness: Rethinking Tasks and Model Evaluation (2023.emnlp-main)
Copied to clipboard
| Challenge: | Language understanding is a multi-faceted cognitive capability, which the Natural Language Processing community has striven to model computationally for decades. |
| Approach: | They propose to rethink what constitutes tasks and model evaluation in NLP and pursue a more holistic view on language, placing trustworthiness at the center. |
| Outcome: | The proposed models are based on generative models and are being deployed in more real-world scenarios, including previously unforeseen zero-shot setups. |
Vicinal Risk Minimization for Few-Shot Cross-lingual Transfer in Abusive Language Detection (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for few-shot cross-lingual transfer learning are limited in target languages due to the scarcity of resources. |
| Approach: | They propose a method which interpolates pairs of instances based on the angle of their representations and propose augmentation methods to enhance few-shot cross-lingual abusive language detection. |
| Outcome: | The proposed method improves few-shot cross-lingual abusive language detection in seven languages typologically distinct from English and three different domains. |
Make Every Letter Count: Building Dialect Variation Dictionaries from Monolingual Corpora (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Dialects exhibit a substantial degree of variation due to the lack of a standard orthography . however, the ability of Large Language Models (LLMs) to process dialects remains understudied . |
| Approach: | They propose a framework for creating dialect variation dictionaries from monolingual data . they use a dataset to examine how well LLMs can judge Bavarian terms as dialect translations . |
| Outcome: | The proposed framework can judge dialects as dialect translations, inflected variants or unrelated forms of a given German lemma. |
How to (Properly) Evaluate Cross-Lingual Word Embeddings: On Strong Baselines, Comparative Analyses, and Some Misconceptions (P19-1)
Copied to clipboard
| Challenge: | Cross-lingual word embeddings (CLEs) are used for downstream NLP tasks . CLEs are based on bilingual lexicon induction (BLI) evaluations vary greatly, hindering ability to interpret performance and properties of different CLE models. |
| Approach: | They evaluate CLE models for a large number of language pairs on bilingual lexicon induction and three downstream tasks. |
| Outcome: | The proposed model performance is based on supervised and unsupervised models on bilingual lexicon induction and three downstream tasks. |
Evaluating Large Language Models for Cross-Lingual Retrieval (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have been evaluated as second-stage reranking models for monolingual IR, but a systematic comparison is lacking for cross-lingual reranked IR. |
| Approach: | They propose to use machine translation to evaluate rerankers in cross-lingual IR . they find that LLMs perform better than LLM-based reranked models . |
| Outcome: | The proposed model improves cross-lingual IR but relies on machine translation for the first stage. |
Cross-Dialect Information Retrieval: Information Access in Low-Resource and High-Variance Languages (2025.coling-main)
Copied to clipboard
| Challenge: | lexical gaps between dialects in cross-lingual information retrieval (CLIR) are caused by orthographic variations and different regional expressions. |
| Approach: | They propose a dataset that consists of seven German dialects extracted from Wikipedia. |
| Outcome: | The proposed dataset consists of seven German dialects extracted from Wikipedia. |
Towards Instance-Level Parser Selection for Cross-Lingual Transfer of Dependency Parsers (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods of cross-lingual parser transfer focus on predicting the best parsers for a low-resource target language globally. |
| Approach: | They propose a cross-lingual parser transfer paradigm that uses instance-level parsers to predict the best parsing for a target language at treebank level. |
| Outcome: | The proposed model outperforms existing models on 13/20 and 14/20 test languages. |
Probing Pretrained Language Models for Lexical Semantics (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have focused on morphosyntactic, semantic, and world knowledge, but it remains unclear to what extent LMs derive lexical type-level knowledge from words in context. |
| Approach: | They propose to use multilingual and monolingual LMs to extract lexical type-level knowledge from words in context. |
| Outcome: | The proposed models perform well across six typologically diverse languages and five lexical tasks. |