A Dutch BERT model encodes gender linearly by epoch 20 but does not dynamically update its representations when explicit female cues contradict learned stereotypical associations in short sentence templates.
hub
Patrick Wilhelm, Thorsten Wittkopp, and Odej Kao
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
hub tools
citation-role summary
method 1
citation-polarity summary
years
2026 11roles
method 1polarities
use method 1representative citing papers
Masked-position MLM plus JEPA latent prediction outperforms MLM-only pretraining on 10-11 of 16 downstream tasks for 35M-150M protein models while JEPA alone fails.
HyperTransport amortizes activation steering for T2I models via a hypernetwork that predicts intervention parameters from CLIP embeddings, delivering 3600-7000x speedup and matching per-concept baselines on 167 unseen concepts.
NorBERTo, a ModernBERT encoder trained on the largest open Portuguese corpus of 331B tokens, reports top encoder results on several PLUE and ASSIN 2 tasks.
Dual Triangle Attention achieves effective bidirectional attention with built-in positional inductive bias via dual triangular masks, outperforming standard bidirectional attention on position-sensitive tasks and showing strong masked language modeling results with or without positional embeddings.
GLiGuard is a compact schema-conditioned bidirectional encoder that matches 7B-27B guard models on safety benchmarks while delivering up to 16x higher throughput and 17x lower latency.
Synthetic reward hacking data does not capture natural hacking behaviors in code generation RL, causing monitors trained on it to generalize poorly compared to those trained on in-the-wild trajectories.
A novel supervised predictor modeling semantic relationships among question, retrieved passages, and generated answer best forecasts when RAG improves QA performance.
RRK compresses documents to multi-token embeddings for efficient listwise reranking, enabling an 8B model to achieve 3x-18x speedups over smaller models with comparable or better effectiveness.
Augmenting commonsense knowledge corpora with negation produces over 2M new triples that benefit LLM negation understanding when used for pre-training.
Zero-shot GPT-OSS detects depression from 1,108 primary care encounter transcripts with AUPRC 0.51 and AUROC 0.77, with meaningful signals in the first 128 patient tokens and added value from dyadic mirroring.
citing papers explorer
-
Is She Even Relevant? When BERT Ignores Explicit Gender Cues
A Dutch BERT model encodes gender linearly by epoch 20 but does not dynamically update its representations when explicit female cues contradict learned stereotypical associations in short sentence templates.
-
ProteinJEPA: Latent prediction complements protein language models
Masked-position MLM plus JEPA latent prediction outperforms MLM-only pretraining on 10-11 of 16 downstream tasks for 35M-150M protein models while JEPA alone fails.
-
HyperTransport: Amortized Conditioning of T2I Generative Models
HyperTransport amortizes activation steering for T2I models via a hypernetwork that predicts intervention parameters from CLIP embeddings, delivering 3600-7000x speedup and matching per-concept baselines on 167 unseen concepts.
-
NorBERTo: A ModernBERT Model Trained for Portuguese with 331 Billion Tokens Corpus
NorBERTo, a ModernBERT encoder trained on the largest open Portuguese corpus of 331B tokens, reports top encoder results on several PLUE and ASSIN 2 tasks.
-
Dual Triangle Attention: Effective Bidirectional Attention Without Positional Embeddings
Dual Triangle Attention achieves effective bidirectional attention with built-in positional inductive bias via dual triangular masks, outperforming standard bidirectional attention on position-sensitive tasks and showing strong masked language modeling results with or without positional embeddings.
-
GLiGuard: Schema-Conditioned Classification for LLM Safeguard
GLiGuard is a compact schema-conditioned bidirectional encoder that matches 7B-27B guard models on safety benchmarks while delivering up to 16x higher throughput and 17x lower latency.
-
Do Synthetic Trajectories Reflect Real Reward Hacking? A Systematic Study on Monitoring In-the-Wild Hacking in Code Generation
Synthetic reward hacking data does not capture natural hacking behaviors in code generation RL, causing monitors trained on it to generalize poorly compared to those trained on in-the-wild trajectories.
-
Rag Performance Prediction for Question Answering
A novel supervised predictor modeling semantic relationships among question, retrieved passages, and generated answer best forecasts when RAG improves QA performance.
-
Efficient Listwise Reranking with Compressed Document Representations
RRK compresses documents to multi-token embeddings for efficient listwise reranking, enabling an 8B model to achieve 3x-18x speedups over smaller models with comparable or better effectiveness.
-
Commonsense Knowledge with Negation: A Resource to Enhance Negation Understanding
Augmenting commonsense knowledge corpora with negation produces over 2M new triples that benefit LLM negation understanding when used for pre-training.
-
Depression Detection at the Point of Care: Automated Analysis of Linguistic Signals from Routine Primary Care Encounters
Zero-shot GPT-OSS detects depression from 1,108 primary care encounter transcripts with AUPRC 0.51 and AUROC 0.77, with meaningful signals in the first 128 patient tokens and added value from dyadic mirroring.