Natural Language Processing

COPPER enhances LLM-based multi-agent collaboration via a self-reflection mechanism and counterfactual PPO. It improves reflection quality, alleviates credit assignment issues, and shows strong perfo…

Reference Trustable Decoding (RTD) revolutionizes large language model adaptation by offering a training-free method, enabling efficient and cost-effective task adaptation without parameter adjustment…

Cross-Layer Attention (CLA) shrinks Transformer Key-Value cache 2x, improving LLMs’ memory efficiency without accuracy loss.

RISE: Recursive Introspection teaches LLMs to iteratively improve their responses, enabling self-correction and enhanced performance on challenging reasoning tasks.

REBORN: An iterative training framework significantly improves unsupervised ASR by learning optimal speech segment boundaries using reinforcement learning, outperforming existing methods.

Overparameterized neural networks surprisingly recover from catastrophic interference when trained cyclically on repeated data sequences, exhibiting anticipatory knowledge reactivation.

Boosting large language model performance after knowledge editing: A new method (D4S) minimizes model damage by regulating the explosive growth of parameter layers, enabling multiple effective edits.

New surrogate loss functions for learning-to-defer achieve Bayes-consistency, realizable H-consistency, and H-consistency bounds simultaneously, resolving open questions and improving L2D performance.

RankRAG: One LLM, dual-purpose instruction-tuning for superior RAG!

Rainbow Teaming: a novel black-box approach generates diverse adversarial prompts to enhance LLM robustness and safety, achieving over 90% attack success rate across various models.

QUEST, a novel Metropolis-Hastings sampling method, generates high-quality & diverse machine translations by using quality metrics as energy functions, overcoming limitations of likelihood-based and r…

Researchers developed a query-based attack that generates adversarial prompts, fooling language models into producing harmful outputs with significantly higher success rates than previous methods, eff…

QuaRot: Revolutionizing 4-bit LLM inference with lossless quantization via rotation!

Weakly supervised strong models outperform weak models; this gain is precisely quantified by the strong model’s misfit error on weak labels.

New APC metric precisely quantifies & optimizes global faithfulness in persona-driven role-playing, offering a fine-grained, explainable evaluation and improving AI character consistency.

QuanTA: Quantum-inspired Tensor Adaptation efficiently fine-tunes LLMs with high-rank updates, surpassing low-rank methods like LoRA for complex tasks while minimizing additional parameters.

QBB: A novel post-training quantization method for LLMs dramatically improves efficiency by replacing multiplications with summations, achieving state-of-the-art results with minimal accuracy loss.

Transformers excel at in-context learning (ICL), solving new tasks with just prompts. This paper provides a mathematical explanation, showing how transformers use multi-concept word semantics to achie…

RLHF’s overoptimization problem is mitigated by RPO, a novel algorithm that uses SFT loss as an implicit adversarial regularizer, ensuring efficient and effective LLM alignment.

ProtGO: A novel unified framework integrating protein sequence, structure & function for superior representation learning, significantly outperforming current methods.