(G)I-DLE: Generative Inference via Distribution-preserving Logit Exclusion with KL Divergence Minimization for Constrained Decoding

• URL: http://arxiv.org/abs/2503.18050v1
• Date: Sun, 23 Mar 2025 12:37:14 GMT
• Title: (G)I-DLE: Generative Inference via Distribution-preserving Logit Exclusion with KL Divergence Minimization for Constrained Decoding
• Abstract summary: (G)I-DLE is a new approach to constrained decoding that leverages KL divergence to preserve the intrinsic conditional probability distribution of autoregressive language models.<n>We validate our method on the K2-Eval dataset, specifically designed to assess Korean language fluency.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We propose (G)I-DLE, a new approach to constrained decoding that leverages KL divergence minimization to preserve the intrinsic conditional probability distribution of autoregressive language models while excluding undesirable tokens. Unlike conventional methods that naively set banned tokens' logits to $-\infty$, which can distort the conversion from raw logits to posterior probabilities and increase output variance, (G)I-DLE re-normalizes the allowed token probabilities to minimize such distortion. We validate our method on the K2-Eval dataset, specifically designed to assess Korean language fluency, logical reasoning, and cultural appropriateness. Experimental results on Qwen2.5 models (ranging from 1.5B to 14B) demonstrate that G-IDLE not only boosts mean evaluation scores but also substantially reduces the variance of output quality.

Related papers

• Breaking the Likelihood Trap: Variance-Calibrated Modulation for Large Language Model Decoding [4.534022278426311]
  Variance-Calibrated Modulation (VCM) is a training-free pre-decoding intervention that reshapes the probability distribution before truncation.<n>Across open-ended generation, factual QA, and mathematical reasoning, VCM consistently mitigates the likelihood trap.
  arXiv  Detail & Related papers  (2026-06-21T14:04:51Z)
• Locally Confident, Globally Stuck: The Quality-Exploration Dilemma in Diffusion Language Models [52.61023005303122]
  We show that low-confidence remasking improves a proxy for quality while constraining the entropy of the induced sequence distribution.<n>We develop a simple Independent--Hastings sampler that approximately targets this distribution during decoding.
  arXiv  Detail & Related papers  (2026-04-01T02:01:30Z)
• Fair regression under localized demographic parity constraints [9.082079838093371]
  We introduce a novel ($ell$, Z)-fair predictor, which imposes groupwise CDF constraints of the form F f |S=s (z m ) = $ell$ m for prescribed pairs.<n> Experiments on synthetic and real datasets illustrate an interpretable fairness-accuracy trade-off, enabling targeted corrections at decision-relevant quantiles or thresholds.
  arXiv  Detail & Related papers  (2026-03-26T09:21:05Z)
• Improving Search Agent with One Line of Code [68.58667107354253]
  Tool-based Agentic Reinforcement Learning (TARL) has emerged as a promising paradigm for training search agents to interact with external tools.<n>We propose textbfSearch textbfAgent textbfPolicy textbfOptimization (textbfSAPO), which stabilizes training via a conditional token-level KL constraint.
  arXiv  Detail & Related papers  (2026-03-10T04:07:39Z)
• Overconfident Errors Need Stronger Correction: Asymmetric Confidence Penalties for Reinforcement Learning [17.384089089363382]
  We identify a root cause that existing methods overlook: the uniform penalization of errors.<n>Current approaches treat all incorrect rollouts within a group identically.<n>We propose the Asymmetric Confidence-aware Error Penalty (ACE)
  arXiv  Detail & Related papers  (2026-02-24T22:46:43Z)
• Bulk-Calibrated Credal Ambiguity Sets: Fast, Tractable Decision Making under Out-of-Sample Contamination [8.826173150779145]
  Distributionally robust optimisation (DRO) minimises the worst-case expected loss over an ambiguity set.<n>We show how IP credal sets translate into DRO objectives with interpretable tolerance levels.
  arXiv  Detail & Related papers  (2026-01-29T06:37:36Z)
• LIME-LLM: Probing Models with Fluent Counterfactuals, Not Broken Text [7.194073942393882]
  We introduce LIME-LLM, a framework that replaces random noise with hypothesis-driven, controlled perturbations.<n> Empirical results demonstrate that LIME-LLM establishes a new benchmark for black-box explainability.
  arXiv  Detail & Related papers  (2026-01-16T19:55:06Z)
• Derailing Non-Answers via Logit Suppression at Output Subspace Boundaries in RLHF-Aligned Language Models [0.0]
  We introduce a method to reduce refusal rates of large language models (LLMs) on sensitive content without modifying model weights or prompts.<n>Motivated by the observation that refusals in certain models were often preceded by the specific token sequence of a token.<n>Our findings suggest that refusal behaviors can be circumvented by blocking refusal subspaces at specific points in the generation process.
  arXiv  Detail & Related papers  (2025-05-28T20:25:24Z)
• Normalized Attention Guidance: Universal Negative Guidance for Diffusion Models [57.20761595019967]
  We present Normalized Attention Guidance (NAG), an efficient, training-free mechanism that applies extrapolation in attention space with L1-based normalization and refinement.<n>NAG restores effective negative guidance where CFG collapses while maintaining fidelity.<n>NAG generalizes across architectures (UNet, DiT), sampling regimes (few-step, multi-step), and modalities (image, video)
  arXiv  Detail & Related papers  (2025-05-27T13:30:46Z)
• Conditional Diffusion Models with Classifier-Free Gibbs-like Guidance [19.83064246586143]
  CFG is a technique for improving conditional diffusion models by linearly combining the outputs of conditional and unconditional denoisers.<n>While CFG enhances visual quality and improves alignment with prompts, it often reduces sample diversity.<n>We propose a Gibbs-like sampling procedure to draw samples from the desired tilted distribution.
  arXiv  Detail & Related papers  (2025-05-27T12:27:33Z)
• Semantic Probabilistic Control of Language Models [27.866585120174484]
  Semantic control involves steering LM generations towards satisfying subtle non-lexical constraints.<n>We leverage a verifier's information gradient to efficiently reason over all generations that satisfy the target attribute.<n>We evaluate the effectiveness of our approach in controlling the toxicity, sentiment, and topic-adherence of LMs yielding generations satisfying the constraint with high probability.
  arXiv  Detail & Related papers  (2025-05-04T01:21:28Z)
• Fast Controlled Generation from Language Models with Adaptive Weighted Rejection Sampling [90.86991492288487]
  evaluating constraint on every token can be prohibitively expensive. LCD can distort the global distribution over strings, sampling tokens based only on local information. We show that our approach is superior to state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-04-07T18:30:18Z)
• Contrastive CFG: Improving CFG in Diffusion Models by Contrasting Positive and Negative Concepts [55.298031232672734]
  As-Free Guidance (CFG) has proven effective in conditional diffusion model sampling for improved condition alignment. We present a novel method to enhance negative CFG guidance using contrastive loss.
  arXiv  Detail & Related papers  (2024-11-26T03:29:27Z)
• Rectified Diffusion Guidance for Conditional Generation [62.00207951161297]
  We revisit the theory behind CFG and rigorously confirm that the improper configuration of the combination coefficients (i.e., the widely used summing-to-one version) brings about expectation shift of the generative distribution. We propose ReCFG with a relaxation on the guidance coefficients such that denoising with ReCFG strictly aligns with the diffusion theory. That way the rectified coefficients can be readily pre-computed via traversing the observed data, leaving the sampling speed barely affected.
  arXiv  Detail & Related papers  (2024-10-24T13:41:32Z)
• Dirichlet-Based Prediction Calibration for Learning with Noisy Labels [40.78497779769083]
  Learning with noisy labels can significantly hinder the generalization performance of deep neural networks (DNNs) Existing approaches address this issue through loss correction or example selection methods. We propose the textitDirichlet-based Prediction (DPC) method as a solution.
  arXiv  Detail & Related papers  (2024-01-13T12:33:04Z)
• A Pseudo-Semantic Loss for Autoregressive Models with Logical Constraints [87.08677547257733]
  Neuro-symbolic AI bridges the gap between purely symbolic and neural approaches to learning. We show how to maximize the likelihood of a symbolic constraint w.r.t the neural network's output distribution. We also evaluate our approach on Sudoku and shortest-path prediction cast as autoregressive generation.
  arXiv  Detail & Related papers  (2023-12-06T20:58:07Z)
• How to train your VAE [0.0]
  Variational Autoencoders (VAEs) have become a cornerstone in generative modeling and representation learning within machine learning. This paper explores interpreting the Kullback-Leibler (KL) Divergence, a critical component within the Evidence Lower Bound (ELBO) The proposed method redefines the ELBO with a mixture of Gaussians for the posterior probability, introduces a regularization term, and employs a PatchGAN discriminator to enhance texture realism.
  arXiv  Detail & Related papers  (2023-09-22T19:52:28Z)
• Tailoring Language Generation Models under Total Variation Distance [55.89964205594829]
  The standard paradigm of neural language generation adopts maximum likelihood estimation (MLE) as the optimizing method. We develop practical bounds to apply it to language generation. We introduce the TaiLr objective that balances the tradeoff of estimating TVD.
  arXiv  Detail & Related papers  (2023-02-26T16:32:52Z)
• Regularizing Variational Autoencoder with Diversity and Uncertainty Awareness [61.827054365139645]
  Variational Autoencoder (VAE) approximates the posterior of latent variables based on amortized variational inference. We propose an alternative model, DU-VAE, for learning a more Diverse and less Uncertain latent space.
  arXiv  Detail & Related papers  (2021-10-24T07:58:13Z)
• Novelty Detection via Robust Variational Autoencoding [13.664682865991255]
  We propose a new method for novelty detection that can tolerate high corruption of the training points. Our method trains a robust variational autoencoder (VAE), which aims to generate a model for the uncorrupted training points.
  arXiv  Detail & Related papers  (2020-06-09T22:34:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.