| Challenge: | Large language models display undesirable behaviors embedded in their internal representations, undermining fairness, inconsistency drift, and the propagation of unwanted patterns during extended dialogues. |
| Approach: | They propose a pruning-based framework that detects context-aware neuron activations and applies adaptive masking to modulate their influence during generation. |
| Outcome: | The proposed framework detects context-aware neuron activations and applies adaptive masking to modulate their influence during generation. |
Similar Papers
Fair-CCD: Mitigating Bias in Large Language Models for Tabular Classification Through Context-Contrastive Decoding (2026.acl-long)
Copied to clipboard
| Challenge: | Prior work to mitigate fairness issues often employs subjective demonstration selection, leading to low controllability and limited stability across different models and tasks. |
| Approach: | They propose to use in-context learning to insert social biases into large language models to create a structured and controllable representation of the relationship between sensitive attributes and predicted labels. |
| Outcome: | Extensive experiments show that Fair-CCD consistently improves fairness metrics without degrading task accuracy. |
BiasFilter: An Inference-Time Debiasing Framework for Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for debiasing large language models incur high human and computational costs and are limited in their effectiveness. |
| Approach: | They propose a model-agnostic, inference-time debiasing framework that enforces fairness by filtering generation outputs in real time. |
| Outcome: | The proposed framework mitigates social bias across a range of LLMs while preserving overall generation quality. |
FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing prompt-based debiasing methods exhibit instability due to sensitivity to prompt changes . fine-tuning-based techniques incur substantial computational overhead and catastrophic forgetting . |
| Approach: | They propose a debiasing framework that encodes fairness-related features into separable directions in the hidden activation space. |
| Outcome: | The proposed framework performs inference-time debiasing without requiring retraining or prompt design . it detects bias signatures in activations and then computes debiased steering vectors . the proposed framework is available to download in the u.s. |
Toward Inclusive Language Models: Sparsity-Driven Calibration for Systematic and Interpretable Mitigation of Social Biases in LLMs (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a new method to mitigate stereotypical bias in large language models is needed . inherent biases from training on vast Internet datasets can amplify harmful stereotypes . |
| Approach: | They propose a method to identify stereotypical bias in decoder-only transformer models . they apply a localization mechanism that correlates internal activations with a new Context Influence score . |
| Outcome: | The proposed method reduces stereotypical biases on BBQ, StereoSet, and CrowS-Pairs while improving reasoning performance on MMLU by 10%. |
DRPruning: Efficient Large Language Model Pruning through Distributionally Robust Optimization (2025.acl-long)
Copied to clipboard
| Challenge: | Structured pruning reduces model size but often causes uneven degradation across domains, leading to biased performance. |
| Approach: | They propose a method that dynamically adjusts the data distribution during training to restore balanced performance across heterogeneous and multi-tasking data. |
| Outcome: | Experiments in monolingual and multilingual settings show that the proposed method surpasses similarly sized models in pruning and continued pretraining over perplexity, downstream tasks, and instruction tuning. |
IG-Pruning: Input-Guided Block Pruning for Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for efficient inference rely on fixed block masks, which can lead to suboptimal performance. |
| Approach: | They propose an input-aware block-wise pruning method that dynamically selects layer masks at inference time. |
| Outcome: | The proposed method outperforms state-of-the-art static depth pruning methods . it is particularly suitable for resource-constrained deployment scenarios . |
The Impossibility of Fair LLMs (2025.acl-long)
Copied to clipboard
| Challenge: | Existing frameworks for evaluating large language models do not extend to general-purpose AI contexts or are infeasible in practice. |
| Approach: | They analyze a variety of technical fairness frameworks to find inherent challenges . they find that each framework does not logically extend to the general-purpose AI context . |
| Outcome: | The proposed frameworks do not logically extend to the general-purpose AI context or are infeasible in practice due to large amounts of unstructured training data and potential combinations of human populations, use cases, and sensitive attributes. |
Iterative Structured Pruning for Large Language Models with Multi-Domain Calibration (2026.eacl-industry)
Copied to clipboard
| Challenge: | Existing models with unstructured pruning often yield irregular sparsity patterns that necessitate specialized hardware or software support. |
| Approach: | They propose a structured pruning framework that eliminates entire architectural components and maintains compatibility with standard hardware accelerators. |
| Outcome: | The proposed model pruning framework achieves significant compression with minimal performance degradation on multiple models across diverse downstream tasks. |
Structured Pruning for Large Language Models Using Coupled Components Elimination and Minor Fine-tuning (2024.findings-naacl)
Copied to clipboard
| Challenge: | Large language models (LLMs) have demonstrated powerful capabilities in natural language processing, yet their vast number of parameters poses challenges for deployment and inference efficiency. |
| Approach: | They propose a structured pruning algorithm that derives the importance of different components based on intermediate data dependencies and removes coupled components across different layers simultaneously. |
| Outcome: | The proposed algorithm reduces model size and accelerates inference without specialized operators and libraries, while maintaining its utility as versatile problem solvers. |
PDTrim: Targeted Pruning for Prefill-Decode Disaggregation in Inference (2026.acl-long)
Copied to clipboard
| Challenge: | Existing pruning methods ignore prefill-decode (PD) disaggregation in practice. |
| Approach: | They propose a pruning method that is highly integrated with prefill-decode (PD) disaggregation, enabling more precise pruning of blocks. |
| Outcome: | The proposed method achieves strong performance in both PD disaggregation and PD unified settings, and can be extended to other non-block pruning methods. |