Evaluating Evidence Grounding Under User Pressure in Instruction-Tuned Language Models

• URL: http://arxiv.org/abs/2603.20162v1
• Date: Fri, 20 Mar 2026 17:38:23 GMT
• Title: Evaluating Evidence Grounding Under User Pressure in Instruction-Tuned Language Models
• Abstract summary: In contested domains, instruction-tuned language models must balance user-alignment pressures against faithfulness to the in-context evidence.<n>We conduct fine-grained ablations over evidence composition and uncertainty cues across 19 instruction-tuned models spanning 0.27B to 32B parameters.
• Score: 6.431677598656395
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In contested domains, instruction-tuned language models must balance user-alignment pressures against faithfulness to the in-context evidence. To evaluate this tension, we introduce a controlled epistemic-conflict framework grounded in the U.S. National Climate Assessment. We conduct fine-grained ablations over evidence composition and uncertainty cues across 19 instruction-tuned models spanning 0.27B to 32B parameters. Across neutral prompts, richer evidence generally improves evidence-consistent accuracy and ordinal scoring performance. Under user pressure, however, evidence does not reliably prevent user-aligned reversals in this controlled fixed-evidence setting. We report three primary failure modes. First, we identify a negative partial-evidence interaction, where adding epistemic nuance, specifically research gaps, is associated with increased susceptibility to sycophancy in families like Llama-3 and Gemma-3. Second, robustness scales non-monotonically: within some families, certain low-to-mid scale models are especially sensitive to adversarial user pressure. Third, models differ in distributional concentration under conflict: some instruction-tuned models maintain sharply peaked ordinal distributions under pressure, while others are substantially more dispersed; in scale-matched Qwen comparisons, reasoning-distilled variants (DeepSeek-R1-Qwen) exhibit consistently higher dispersion than their instruction-tuned counterparts. These findings suggest that, in a controlled fixed-evidence setting, providing richer in-context evidence alone offers no guarantee against user pressure without explicit training for epistemic integrity.

Related papers

• Does Reasoning Preserve Alignment? On the Trustworthiness of Large Reasoning Models [55.788110316999166]
  We compare reasoning models produced via supervised fine-tuning, RL-based post-training, and distillation against instruction-tuned baselines.<n>We observe that reasoning models often improve on reasoning benchmarks but exhibit alignment regressions.<n>These regressions are consistent with behavioral drift from the instruction-tuned baseline, measured by KL divergence.
  arXiv  Detail & Related papers  (2026-06-09T16:14:27Z)
• Not All Flips Are Conformity: Decomposing Stance Convergence in Multi-Agent LLM Debate [5.650336594658653]
  We show that a conventional answer flip rate conflates three mechanisms: spontaneous instability, stance-induced conformity, and reasoning-induced persuasion.<n>Our three-source decomposition framework isolates each through controlled counterfactual conditions.
  arXiv  Detail & Related papers  (2026-05-30T17:41:11Z)
• Do Language Models Know What Not to Say? Causal Evidence for Statistical Preemption in LLMs [13.891522069967507]
  Construction Grammar proposes statistical preemption: exposure to a conventional form preempts structurally possible but unattested alternatives.<n>We present a computational study that dissociates statistical preemption from the competing entrenchment hypothesis in large language models.
  arXiv  Detail & Related papers  (2026-05-21T21:06:43Z)
• Towards Trustworthy Depression Estimation via Disentangled Evidential Learning [50.22167852149165]
  We propose EviDep, an evidential learning framework that jointly quantifies depression severity.<n>EviDep enforces strict information integrity to guarantee robust evidence synthesis.<n>It achieves state-of-the-art predictive accuracy and superior uncertainty calibration, delivering a robust fail-safe mechanism for trustworthy clinical screening.
  arXiv  Detail & Related papers  (2026-04-17T13:27:11Z)
• Beyond Social Pressure: Benchmarking Epistemic Attack in Large Language Models [0.07812854697536452]
  We develop a diagnostic benchmark for evaluating textitepistemic attack<n> prompts challenge the legitimacy of knowledge, values, or identity rather than simply opposing a previous answer.
  arXiv  Detail & Related papers  (2026-04-09T03:14:30Z)
• Pressure, What Pressure? Sycophancy Disentanglement in Language Models via Reward Decomposition [4.5328321409938095]
  We operationalise sycophancy through formal definitions of pressure independence and evidence responsiveness.<n>We propose the first approach to sycophancy reduction via reward decomposition.
  arXiv  Detail & Related papers  (2026-04-07T00:28:17Z)
• On the Structural Non-Preservation of Epistemic Behaviour under Policy Transformation [51.56484100374058]
  We formalise such information-conditioned interaction patterns as behavioural dependency.<n>This induces a probe-relative notion of $$-behavioural equivalence and a within-policy behavioural distance.<n>Results identify structural conditions under which probe-conditioned behavioural separation is not preserved under common policy transformations.
  arXiv  Detail & Related papers  (2026-02-24T22:55:21Z)
• Same Answer, Different Representations: Hidden instability in VLMs [65.36933543377346]
  We introduce a representation-aware and frequency-aware evaluation framework that measures internal embedding drift, spectral sensitivity, and structural smoothness.<n>We apply this framework to modern Vision Language Models (VLMs) across the SEEDBench, MMMU, and POPE datasets.
  arXiv  Detail & Related papers  (2026-02-06T12:24:26Z)
• Decomposed Prompting Does Not Fix Knowledge Gaps, But Helps Models Say "I Don't Know" [47.930782177987446]
  Large language models often struggle to recognize their knowledge limits in closed-book question answering, leading to confident hallucinations.<n>We evaluate three task-equivalent prompting regimes: Direct, Assistive, and Incremental, across different model scales and multi-hop QA benchmarks.<n>Because factual knowledge is stable while hallucinations are agreement, cross-regime provides a precise signal of internal uncertainty.
  arXiv  Detail & Related papers  (2026-02-04T18:39:58Z)
• Compressed Causal Reasoning: Quantization and GraphRAG Effects on Interventional and Counterfactual Accuracy [0.0]
  This study systematically evaluate quantization effects across all three levels of Pearls Causal Ladder.<n>We find that rung level accuracy in Llama 3 8B remains broadly stable under quantization, with NF4 showing less than one percent overall degradation.<n>Experiments on the CRASS benchmark show near identical performance across precisions, indicating that existing commonsense counterfactual datasets lack the structural sensitivity needed to reveal quantization induced reasoning drift.
  arXiv  Detail & Related papers  (2025-12-13T17:54:15Z)
• Counterfactual Reward Model Training for Bias Mitigation in Multimodal Reinforcement Learning [0.5204229323525671]
  We present a counterfactual reward model that introduces causal inference with multimodal representation learning to provide an unsupervised, bias-resilient reward signal.<n>We evaluated the framework on a multimodal fake versus true news dataset, which exhibits framing bias, class imbalance, and distributional drift.<n>The resulting system achieved an accuracy of 89.12% in fake news detection, outperforming the baseline reward models.
  arXiv  Detail & Related papers  (2025-08-27T04:54:33Z)
• Sycophancy under Pressure: Evaluating and Mitigating Sycophantic Bias via Adversarial Dialogues in Scientific QA [36.21980066799023]
  sycophancy is the tendency to align with user beliefs regardless of correctness.<n>Despite its importance, sycophancy remains underexamined in factual question answering contexts.<n>We introduce a unified evaluation framework to quantify the impact of sycophantic context on model behavior.
  arXiv  Detail & Related papers  (2025-08-19T11:30:52Z)
• Reasoning Models Are More Easily Gaslighted Than You Think [85.84943447589511]
  We evaluate three state-of-the-art reasoning models, including OpenAI's o4-mini, Claude-3.7-Sonnet and Gemini-2.5-Flash.<n>Our evaluation reveals significant accuracy drops following gaslighting negation prompts.<n>We introduce GaslightingBench-R, a new diagnostic benchmark designed to evaluate reasoning models' susceptibility to defend their belief.
  arXiv  Detail & Related papers  (2025-06-11T12:52:25Z)
• Explicit Tradeoffs between Adversarial and Natural Distributional Robustness [48.44639585732391]
  In practice, models need to enjoy both types of robustness to ensure reliability. In this work, we show that in fact, explicit tradeoffs exist between adversarial and natural distributional robustness.
  arXiv  Detail & Related papers  (2022-09-15T19:58:01Z)
• Fundamental Tradeoffs between Invariance and Sensitivity to Adversarial Perturbations [65.05561023880351]
  Adversarial examples are malicious inputs crafted to induce misclassification. This paper studies a complementary failure mode, invariance-based adversarial examples. We show that defenses against sensitivity-based attacks actively harm a model's accuracy on invariance-based attacks.
  arXiv  Detail & Related papers  (2020-02-11T18:50:23Z)

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.