Papers by Kristian Kersting

10 papers
Multilingual Text-to-Image Generation Magnifies Gender Stereotypes (2025.acl-long)

Copied to clipboard

Challenge: Text-to-image (T2I) generation models have great results in image quality, flexibility, and text alignment, but they suffer from substantial gender bias.
Approach: They propose a benchmark to study gender bias in multilingual T2I models . they use multilingual prompts to account for grammatical differences influencing gender .
Outcome: The proposed benchmark shows strong gender biases and language-specific differences across models.
Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning (2026.acl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) are stateless and limited by a finite context window, preventing them from maintaining knowledge across long conversations or evolving tasks.
Approach: They propose a reinforcement learning framework that empowers LLMs to actively manage external memory through two specialized agents.
Outcome: The proposed framework outperforms baselines and benchmarks across diverse question types, three benchmarks, and multiple model scales.
Speaking Multiple Languages Affects the Moral Bias of Language Models (2023.findings-acl)

Copied to clipboard

Challenge: Pre-trained multilingual language models are often better on English than other languages . however, they are trained on varying amounts of data for each language .
Approach: They apply the MORALDIRECTION framework to multilingual models and analyse their results . they find that PMLMs encode differing moral biases, but these do not correspond to cultural differences or commonalities in human opinions.
Outcome: The proposed model captures moral norms from English and imposes them on other languages.
Adaptable Adapters (2022.naacl-main)

Copied to clipboard

Challenge: Existing work uses the same adapter architecture for every dataset regardless of the properties of the dataset or the amount of training data.
Approach: They propose to use adaptable adapters to finetune lightweight neural network layers on top of pretrained weights.
Outcome: The proposed adapters achieve on-par performances with the standard adapter architecture while using a considerably smaller number of adapter layers.
Divergent Token Metrics: Measuring degradation to prune away LLM components – and optimize quantization (2024.naacl-long)

Copied to clipboard

Challenge: Large Language Models (LLMs) have reshaped natural language processing with impressive capabilities, but their ever-increasing size has raised concerns about their effective deployment and the need for LLM compression.
Approach: This study introduces the Divergent Token Metrics (DTMs) that measure token divergences that allow deeper insights into the subtleties of model compression.
Outcome: The proposed measures can identify outliers and improve performance in the sparseness of the LLMs.
Judging Quality Across Languages: A Multilingual Approach to Pretraining Data Filtering with Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Existing open-source multilingual datasets rely on heuristic filtering methods restricting both their cross-lingual transferability and scalability.
Approach: They propose a systematic approach that curates diverse and high-quality multilingual data at scale while significantly reducing computational demands.
Outcome: Evaluated empirically across 35 languages, the proposed approach outperforms current heuristic filtering methods like Fineweb2 and improves model training quality and retention rates.
T-FREE: Subword Tokenizer-Free Generative LLMs via Sparse Representations for Memory-Efficient Embeddings (2024.emnlp-main)

Copied to clipboard

Challenge: Tokenizers are crucial for encoding information in Large Language Models, but their development has stagnated.
Approach: They propose a tokenizer that embeds words through sparse activation patterns over character triplets . they show competitive downstream performance with a parameter reduction of more than 85% .
Outcome: The proposed approach achieves competitive downstream performance with a parameter reduction of more than 85% on embedding layers.
SLR: Automated Synthesis for Scalable Logical Reasoning (2026.acl-long)

Copied to clipboard

Challenge: Existing benchmarks intended to evaluate reasoning capabilities emphasize deductive reasoning, where conclusions necessarily follow from given premises.
Approach: They propose an end-to-end framework for systematic evaluation and training of Large Language Models via Scalable Logical Reasoning.
Outcome: The proposed framework doubles Llama-3-8B accuracy on SLR-Bench, achieving parity with Gemini-Flash-Thinking at a fraction of computational cost.
METok: Multi-Stage Event-based Token Compression for Efficient Long Video Understanding (2025.emnlp-main)

Copied to clipboard

Challenge: Recent advances in Video Large Language Models (VLLMs) have significantly enhanced their ability to understand video content.
Approach: They propose a training-free, Multi-stage Event-based Token compression framework that eliminates redundant visual tokens across three critical stages .
Outcome: The proposed framework reduces FLOPs and KV Cache memory consumption while maintaining comparable or even superior accuracy.
STRICTA: Structured Reasoning in Critical Text Assessment for Peer Review and Beyond (2025.acl-long)

Copied to clipboard

Challenge: Existing work treats critical text assessment as a black box problem, limiting interpretability and human-AI collaboration.
Approach: They propose a framework to model critical text assessment as an explicit, step-wise reasoning process.
Outcome: The proposed framework breaks down assessment into a graph of interconnected reasoning steps drawing on causality theory.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations