Ubiquitous Knowledge Processing Lab

university

https://www.informatik.tu-darmstadt.de/ukp

UKPLab

UKPLab

Activity Feed Request to join this org

AI & ML interests

The Ubiquitous Knowledge Processing Lab researches natural language processing, text mining, eLearning, and digital humanities.

Recent Activity

imbesat-rizvi updated a model about 14 hours ago

UKPLab/Qwen2.5-3b-spare-prm-math

imbesat-rizvi published a model about 15 hours ago

UKPLab/Qwen2.5-3b-spare-prm-math

haritzpuerto authored a paper about 1 month ago

Leaky Thoughts: Large Reasoning Models Are Not Private Thinkers

View all activity

imbesat-rizvi

updated a model about 14 hours ago

UKPLab/Qwen2.5-3b-spare-prm-math

Text Generation • 3B • Updated about 14 hours ago

imbesat-rizvi

published a model about 15 hours ago

UKPLab/Qwen2.5-3b-spare-prm-math

Text Generation • 3B • Updated about 14 hours ago

haritzpuerto

authored 2 papers about 1 month ago

Leaky Thoughts: Large Reasoning Models Are Not Private Thinkers

Paper • 2506.15674 • Published Jun 18

C-SEO Bench: Does Conversational SEO Work?

Paper • 2506.11097 • Published Jun 6

imbesat-rizvi

updated a model 2 months ago

UKPLab/Llama-3-8b-spare-prm-math

Text Generation • 8B • Updated Jun 19 • 11

imbesat-rizvi

updated a dataset 3 months ago

UKPLab/sparp

Viewer • Updated Jun 5 • 417k • 36 • 1

imbesat-rizvi

published a model 3 months ago

UKPLab/Llama-3-8b-spare-prm-math

Text Generation • 8B • Updated Jun 19 • 11

iNeil77

authored a paper 3 months ago

Massively Multilingual Adaptation of Large Language Models Using Bilingual Translation Data

Paper • 2506.00469 • Published May 31 • 2

haritzpuerto

posted an update 3 months ago

Post

325

📜 Accepted at ACL 2025! Fine-Tuning on Diverse Reasoning Chains Drives Within-Inference CoT Refinement in LLMs
We propose to fine-tune LLMs to generate diverse chains of thought (DCoT) in a single inference step. This enables within-inference refinement of the cots, no external feedback needed!
🔗 https://arxiv.org/abs/2407.03181

Rachneet

updated a dataset 3 months ago

UKPLab/HaluQuestQA

Viewer • Updated May 28 • 698 • 12

Rachneet

published a dataset 3 months ago

UKPLab/HaluQuestQA

Viewer • Updated May 28 • 698 • 12

timbmg

updated a dataset 4 months ago

UKPLab/PeerQA

Updated May 7 • 114 • 2

sukannya

authored 2 papers 4 months ago

CVQA: Culturally-diverse Multilingual Visual Question Answering Benchmark

Paper • 2406.05967 • Published Jun 10, 2024 • 6

LazyReview A Dataset for Uncovering Lazy Thinking in NLP Peer Reviews

Paper • 2504.11042 • Published Apr 15 • 4

iNeil77

authored 3 papers 5 months ago

EMMA-500: Enhancing Massively Multilingual Adaptation of Large Language Models

Paper • 2409.17892 • Published Sep 26, 2024 • 2

ObscuraCoder: Powering Efficient Code LM Pre-Training Via Obfuscation Grounding

Paper • 2504.00019 • Published Mar 27

IRCoder: Intermediate Representations Make Language Models Robust Multilingual Code Generators

Paper • 2403.03894 • Published Mar 6, 2024

Justus-Jonas

authored 3 papers 5 months ago

Triple-Encoders: Representations That Fire Together, Wire Together

Paper • 2402.12332 • Published Feb 19, 2024 • 2

Imagination is All You Need! Curved Contrastive Learning for Abstract Sequence Modeling Utilized on Long Short-Term Dialogue Planning

Paper • 2211.07591 • Published Nov 14, 2022

GRITHopper: Decomposition-Free Multi-Hop Dense Retrieval

Paper • 2503.07519 • Published Mar 10