Subword tokenization impairs phonological knowledge encoding in LMs, but an IPA-based fine-tuning method restores it with minimal impact on other capabilities.
Rephrase and respond: Let large language models ask better questions for themselves.CoRR, abs/2311.04205
10 Pith papers cite this work, alongside 12 external citations. Polarity classification is still indexing.
10
Pith papers citing it
12
external citations · external index
verdicts
UNVERDICTED 10representative citing papers
Fine-tuned ModernBERT-family encoders match LLM judges on F1, false negative rate, and precision-recall for harmful output detection across adversarial datasets and attack types while promising lower cost and latency.
SynLearner lets LLMs improve synthetic data generation on later tasks in a stream by learning reusable patterns and balancing quality with diversity from feedback on earlier tasks.
Proposes nearly balanced TCARDs that minimize the first two generalized word-length pattern components, defines Φ_BCD criterion linked to classical optimality, and constructs designs via coordinate exchange with simulation-calibrated weights for LLM prompt engineering.
D2R-RAG diagnoses RAG failures via observable signals and applies adaptive repairs under latency and VRAM limits, reporting improved reliability and accuracy-efficiency trade-offs on FEVER and HotpotQA.
A taxonomy-guided RAG system with LLMs reduces hallucinations and improves migration suggestions for Qiskit code compared to unconstrained retrieval.
Authors develop a multi-dimensional neuron screening framework and adaptive masking method to causally validate and steer emotion and rhetoric neurons in LLMs, with experiments on five datasets.
A new prompting framework called Thought Graph Traversal combined with reasoning budget forcing improves test-time performance of frozen chest X-ray VLLMs on report generation benchmarks.
A systematic survey categorizes prompt engineering methods for LLMs and VLMs by application area, summarizing methodologies, applications, models, datasets, strengths, and limitations for each technique along with a taxonomy and summary table.
This survey frames foundation agents using brain-inspired modular architectures and reviews challenges in evolution, collaboration, and safety.
citing papers explorer
-
How Tokenization Limits Phonological Knowledge Representation in Language Models and How to Improve Them
Subword tokenization impairs phonological knowledge encoding in LMs, but an IPA-based fine-tuning method restores it with minimal impact on other capabilities.
-
Do Encoders Suffice? A Systematic Comparison of Encoder and Decoder Safety Judges for LLM Adversarial Evaluation
Fine-tuned ModernBERT-family encoders match LLM judges on F1, false negative rate, and precision-recall for harmful output detection across adversarial datasets and attack types while promising lower cost and latency.
-
Make LLM Learn to Synthesize from Streaming Experiences through Feedback
SynLearner lets LLMs improve synthetic data generation on later tasks in a stream by learning reusable patterns and balancing quality with diversity from feedback on earlier tasks.
-
TCARD: Nearly Balanced Two-Level Designs with Treatment Cardinality Constraints with an Application to LLM Prompt Engineering
Proposes nearly balanced TCARDs that minimize the first two generalized word-length pattern components, defines Φ_BCD criterion linked to classical optimality, and constructs designs via coordinate exchange with simulation-calibrated weights for LLM prompt engineering.
-
Diagnosing and Repairing Factual Errors in RAG under Budget Constraints
D2R-RAG diagnoses RAG failures via observable signals and applies adaptive repairs under latency and VRAM limits, reporting improved reliability and accuracy-efficiency trade-offs on FEVER and HotpotQA.
-
Qiskit Code Migration with LLMs
A taxonomy-guided RAG system with LLMs reduces hallucinations and improves migration suggestions for Qiskit code compared to unconstrained retrieval.
-
Are Emotion and Rhetoric Neurons in LLM? Neuron Recognition and Adaptive Masking for Emotion-Rhetoric Prediction Steering
Authors develop a multi-dimensional neuron screening framework and adaptive masking method to causally validate and steer emotion and rhetoric neurons in LLMs, with experiments on five datasets.
-
Thought Graph Traversal for Test-time Scaling in Chest X-ray VLLMs
A new prompting framework called Thought Graph Traversal combined with reasoning budget forcing improves test-time performance of frozen chest X-ray VLLMs on report generation benchmarks.
-
A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications
A systematic survey categorizes prompt engineering methods for LLMs and VLMs by application area, summarizing methodologies, applications, models, datasets, strengths, and limitations for each technique along with a taxonomy and summary table.
-
Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems
This survey frames foundation agents using brain-inspired modular architectures and reviews challenges in evolution, collaboration, and safety.