AfrIFact provides a multi-stage fact-checking dataset for ten African languages, exposing gaps in embedding models and LLMs for low-resource cultural and health claims.
Tiny Aya: Bridging scale and multilingual depth,
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
years
2026 7representative citing papers
An ablation study of an RL-based jailbreaker finds the attack succeeds across tested open-weight models and safeguards, but its headline conclusion about dense rewards and long episodes is contradicted by its own data.
CroCo applies English-reward-ranked self-generations for contrastive preference tuning that improves two LLMs on structured and open-ended tasks across 14 languages without language-specific annotations.
Merging any combination of monolingual pre-trained models leads to performance collapse due to interference, indicating that merging flexibility from fine-tuning does not extend to pre-training.
A survey of 232 papers frames the 'last mile' challenge of deploying multilingual language models under edge hardware constraints for Global South communities.
Qwen 3.5 4B reaches 96.60% F1 on merchant information extraction after LoRA fine-tuning, within 0.35 points of the LLaMA 3.1-8B baseline at 96.95%, with smaller models showing usable accuracy-latency trade-offs.
A neuro-symbolic pipeline pairing 4B-parameter LLMs with a symbolic theorem prover delivers competitive accuracy and low content effects on syllogistic reasoning subtasks.
citing papers explorer
-
AfrIFact: Cultural Information Retrieval, Evidence Extraction and Fact Checking for African Languages
AfrIFact provides a multi-stage fact-checking dataset for ten African languages, exposing gaps in embedding models and LLMs for low-resource cultural and health claims.
-
A Systematic Investigation of RL-Jailbreaking in LLMs
An ablation study of an RL-based jailbreaker finds the attack succeeds across tested open-weight models and safeguards, but its headline conclusion about dense rewards and long episodes is contradicted by its own data.
-
CroCo: Cross-Lingual Contrastive Preference Tuning on Self-Generations
CroCo applies English-reward-ranked self-generations for contrastive preference tuning that improves two LLMs on structured and open-ended tasks across 14 languages without language-specific annotations.
-
On the Limits of Model Merging for Multilinguality in Pre-Training
Merging any combination of monolingual pre-trained models leads to performance collapse due to interference, indicating that merging flexibility from fine-tuning does not extend to pre-training.
-
Multilinguality at the Edge: Developing Language Models for the Global South
A survey of 232 papers frames the 'last mile' challenge of deploying multilingual language models under edge hardware constraints for Global South communities.
-
How Small Can You Go? LoRA Fine-Tuning 270M-8B Models for Merchant Information Extraction in Financial Transactions
Qwen 3.5 4B reaches 96.60% F1 on merchant information extraction after LoRA fine-tuning, within 0.35 points of the LLaMA 3.1-8B baseline at 96.95%, with smaller models showing usable accuracy-latency trade-offs.
-
UFAL-CUNI at SemEval-2026 Task 11: An Efficient Modular Neuro-symbolic Method for Syllogistic Reasoning
A neuro-symbolic pipeline pairing 4B-parameter LLMs with a symbolic theorem prover delivers competitive accuracy and low content effects on syllogistic reasoning subtasks.