Uncertainty Unveiled: Can Exposure to More In-context Examples Mitigate Uncertainty for Large Language Models?

• URL: http://arxiv.org/abs/2505.21003v1
• Date: Tue, 27 May 2025 10:36:39 GMT
• Title: Uncertainty Unveiled: Can Exposure to More In-context Examples Mitigate Uncertainty for Large Language Models?
• Abstract summary: We investigate how increased examples influence predictive uncertainty, an essential aspect in trustworthiness.<n>Our results reveal that additional examples reduce total uncertainty in both simple and complex tasks by injecting task-specific knowledge.<n>For complex tasks, these advantages emerge only after addressing the increased noise and uncertainty associated with longer inputs.
• Score: 14.333412847910601
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in handling long sequences have facilitated the exploration of long-context in-context learning (ICL). While much of the existing research emphasizes performance improvements driven by additional in-context examples, the influence on the trustworthiness of generated responses remains underexplored. This paper addresses this gap by investigating how increased examples influence predictive uncertainty, an essential aspect in trustworthiness. We begin by systematically quantifying the uncertainty of ICL with varying shot counts, analyzing the impact of example quantity. Through uncertainty decomposition, we introduce a novel perspective on performance enhancement, with a focus on epistemic uncertainty (EU). Our results reveal that additional examples reduce total uncertainty in both simple and complex tasks by injecting task-specific knowledge, thereby diminishing EU and enhancing performance. For complex tasks, these advantages emerge only after addressing the increased noise and uncertainty associated with longer inputs. Finally, we explore the evolution of internal confidence across layers, unveiling the mechanisms driving the reduction in uncertainty.

Related papers

• Exploring Agentic Tool-Calling Decisions via Uncertainty-Aligned Reinforcement Learning [63.24624171679711]
  Large language model (LLM)-based agents often make suboptimal tool-use decisions.<n>We propose TRUST, which incorporates uncertainty quantification into reward design as a repulsive force for maintaining uncertainty separation.
  arXiv  Detail & Related papers  (2026-06-05T07:08:34Z)
• Perception Without Engagement: Dissecting the Causal Discovery Deficit in LMMs [68.58207076756237]
  We introduce ProCauEval, a perturbation-based evaluation protocol that shifts from outcome assessment to mechanism diagnosis.<n>We find that models faithfully perceive video content yet systematically underexploit it during causal reasoning.<n>We propose Anti-Distillation Policy Optimization (ADPO), a reinforcement learning framework built on negative teacher alignment.
  arXiv  Detail & Related papers  (2026-05-10T08:48:58Z)
• Less Approximates More: Harmonizing Performance and Confidence Faithfulness via Hybrid Post-Training for High-Stakes Tasks [52.81286869496811]
  Large language models are increasingly deployed in high-stakes tasks, where confident yet incorrect inferences may cause real-world harm.<n>We propose to jointly optimize unsupervised Reinforcement Learning from Internal Feedback with reasoning-trace-guided Reasoning Distillation.<n>We introduce HyTuning, a hybrid post-training framework that adaptively reweights RD and RLIF via a PRG-style metric.
  arXiv  Detail & Related papers  (2026-04-09T16:50:11Z)
• Self-Consolidation for Self-Evolving Agents [51.94826934403236]
  Large language model (LLM) agents operate as static systems, lacking the ability to evolve through lifelong interaction.<n>We propose a novel self-evolving framework for LLM agents that introduces a complementary evolution mechanism.
  arXiv  Detail & Related papers  (2026-02-02T11:16:07Z)
• Beware of Reasoning Overconfidence: Pitfalls in the Reasoning Process for Multi-solution Tasks [54.31998314008198]
  Large Language Models (LLMs) excel in reasoning tasks requiring a single correct answer, but they perform poorly in multi-solution tasks.<n>We attribute this limitation to textbfreasoning overconfidence: a tendency to express undue certainty in an incomplete solution set.<n>We propose the textbfcognitive-rigidity hypothesis, which posits that overconfidence arises when the reasoning process prematurely converges on a narrow set of thought paths.
  arXiv  Detail & Related papers  (2025-12-01T14:35:06Z)
• Towards Reliable LLM-based Robot Planning via Combined Uncertainty Estimation [68.106428321492]
  Large language models (LLMs) demonstrate advanced reasoning abilities, enabling robots to understand natural language instructions and generate high-level plans with appropriate grounding.<n>LLMs hallucinations present a significant challenge, often leading to overconfident yet potentially misaligned or unsafe plans.<n>We present Combined Uncertainty estimation for Reliable Embodied planning (CURE), which decomposes the uncertainty into epistemic and intrinsic uncertainty, each estimated separately.
  arXiv  Detail & Related papers  (2025-10-09T10:26:58Z)
• Assessing the Potential for Catastrophic Failure in Dynamic Post-Training Quantization [3.437656066916039]
  Post-training quantization (PTQ) has emerged as an effective tool for reducing the computational complexity and memory usage of a neural network.<n>There is the potential for drastic performance reduction depending upon the distribution of inputs experienced in inference.
  arXiv  Detail & Related papers  (2025-10-02T18:13:06Z)
• Quantifying Self-Awareness of Knowledge in Large Language Models [19.886013407039517]
  Hallucination prediction in large language models is often interpreted as a sign of self-awareness.<n>We argue that such performance can arise from question-side shortcuts rather than true model-side introspection.<n>We introduce SCAO, a method that enhances the use of model-side signals.
  arXiv  Detail & Related papers  (2025-09-18T18:29:14Z)
• Variational Uncertainty Decomposition for In-Context Learning [16.986925734554447]
  We introduce a variational uncertainty decomposition framework for in-context learning.<n>We optimise auxiliary queries as probes to obtain an upper bound to the aleatoric uncertainty of an in-context learning procedure.
  arXiv  Detail & Related papers  (2025-09-02T13:53:09Z)
• The Role of Model Confidence on Bias Effects in Measured Uncertainties [11.314633260055436]
  We find that mitigating prompt-introduced bias improves uncertainty quantification in Visual Question Answering (VQA) tasks.<n>We find that all considered biases induce greater changes in both uncertainties when bias-free model confidence is lower.<n>These distinct effects deepen our understanding of bias mitigation for uncertainty quantification and potentially inform the development of more advanced techniques.
  arXiv  Detail & Related papers  (2025-06-20T03:43:10Z)
• Enhancing Uncertainty Estimation and Interpretability via Bayesian Non-negative Decision Layer [55.66973223528494]
  We develop a Bayesian Non-negative Decision Layer (BNDL), which reformulates deep neural networks as a conditional Bayesian non-negative factor analysis.<n>BNDL can model complex dependencies and provide robust uncertainty estimation.<n>We also offer theoretical guarantees that BNDL can achieve effective disentangled learning.
  arXiv  Detail & Related papers  (2025-05-28T10:23:34Z)
• ConCISE: Confidence-guided Compression in Step-by-step Efficient Reasoning [75.1101108949743]
  Large Reasoning Models (LRMs) perform strongly in complex reasoning tasks via Chain-of-Thought (CoT) prompting.<n>LRMs often suffer from verbose outputs caused by redundant content, increasing computational overhead, and degrading user experience.<n>We propose ConCISE, a framework that simplifies reasoning chains by reinforcing the model's confidence during inference.
  arXiv  Detail & Related papers  (2025-05-08T01:40:40Z)
• A Survey of Efficient Reasoning for Large Reasoning Models: Language, Multimodality, and Beyond [88.5807076505261]
  Large Reasoning Models (LRMs) have demonstrated strong performance gains by scaling up the length of Chain-of-Thought (CoT) reasoning during inference.<n>A growing concern lies in their tendency to produce excessively long reasoning traces.<n>This inefficiency introduces significant challenges for training, inference, and real-world deployment.
  arXiv  Detail & Related papers  (2025-03-27T15:36:30Z)
• ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented Generation [91.20492150248106]
  We investigate the internal mechanisms behind unfaithful generation and identify a subset of mid-to-deep feed-forward networks (FFNs) that are disproportionately activated in such cases.<n>We propose Parametric Knowledge Muting through FFN Suppression (ParamMute), a framework that improves contextual faithfulness by suppressing the activation of unfaithfulness-associated FFNs.<n> Experimental results show that ParamMute significantly enhances faithfulness across both CoFaithfulQA and the established ConFiQA benchmark, achieving substantial reductions in reliance on parametric memory.
  arXiv  Detail & Related papers  (2025-02-21T15:50:41Z)
• Contextual Bandits in Payment Processing: Non-uniform Exploration and Supervised Learning at Adyen [0.0]
  Regression oracles combine non-uniform exploration with supervised learning.<n>We analyze these approaches within a real-world industrial context at Adyen, a large global payments processor.<n>While regression oracles significantly improve performance, they introduce challenges due to rigid algorithmic assumptions.
  arXiv  Detail & Related papers  (2024-11-30T19:45:23Z)
• Ensured: Explanations for Decreasing the Epistemic Uncertainty in Predictions [1.2289361708127877]
  Epistem uncertainty adds a crucial dimension to explanation quality. We introduce new types of explanations that specifically target this uncertainty. We introduce a new metric, ensured ranking, designed to help users identify the most reliable explanations.
  arXiv  Detail & Related papers  (2024-10-07T20:21:51Z)
• Variable-Agnostic Causal Exploration for Reinforcement Learning [56.52768265734155]
  We introduce a novel framework, Variable-Agnostic Causal Exploration for Reinforcement Learning (VACERL) Our approach automatically identifies crucial observation-action steps associated with key variables using attention mechanisms. It constructs the causal graph connecting these steps, which guides the agent towards observation-action pairs with greater causal influence on task completion.
  arXiv  Detail & Related papers  (2024-07-17T09:45:27Z)
• Uncertainty-boosted Robust Video Activity Anticipation [72.14155465769201]
  Video activity anticipation aims to predict what will happen in the future, embracing a broad application prospect ranging from robot vision to autonomous driving. Despite the recent progress, the data uncertainty issue, reflected as the content evolution process and dynamic correlation in event labels, has been somehow ignored. We propose an uncertainty-boosted robust video activity anticipation framework, which generates uncertainty values to indicate the credibility of the anticipation results.
  arXiv  Detail & Related papers  (2024-04-29T12:31:38Z)
• Evidential uncertainty sampling for active learning [4.271684331748043]
  Recent studies in active learning have focused on the decomposition of model uncertainty into reducible and irreducible uncertainties. In this paper, the aim is to simplify the computational process while eliminating the dependence on observations.
  arXiv  Detail & Related papers  (2023-09-21T21:26:50Z)
• Post-Abstention: Towards Reliably Re-Attempting the Abstained Instances in QA [24.572043878403353]
  We present an explorative study on 'Post-Abstention', a task that allows re-attempting the abstained instances with the aim of increasing 'coverage' of the system. Comprehensive experiments on 11 QA datasets show that these methods lead to considerable risk improvements.
  arXiv  Detail & Related papers  (2023-05-02T22:40:41Z)
• Model-Based Uncertainty in Value Functions [89.31922008981735]
  We focus on characterizing the variance over values induced by a distribution over MDPs. Previous work upper bounds the posterior variance over values by solving a so-called uncertainty Bellman equation. We propose a new uncertainty Bellman equation whose solution converges to the true posterior variance over values.
  arXiv  Detail & Related papers  (2023-02-24T09:18:27Z)
• Residual Error: a New Performance Measure for Adversarial Robustness [85.0371352689919]
  A major challenge that limits the wide-spread adoption of deep learning has been their fragility to adversarial attacks. This study presents the concept of residual error, a new performance measure for assessing the adversarial robustness of a deep neural network. Experimental results using the case of image classification demonstrate the effectiveness and efficacy of the proposed residual error metric.
  arXiv  Detail & Related papers  (2021-06-18T16:34:23Z)

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.