# When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysis

arXiv: https://arxiv.org/abs/2606.29251 · HTML: https://mlsanity.com/paper/2606.29251/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Efficient reasoning and agent inference, one of 12 directions in Efficient LLM inference and serving.
- Builds on 25 of its 33 references that are in the map.
- First posted on arXiv 28 Jun 2026.

## Placed in

- Efficient LLM inference and serving / [Efficient reasoning and agent inference](https://mlsanity.com/inference/efficient-reasoning-and-agent-inference/)

## Builds on (most cited first, 25 of 25)

- [FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation](https://mlsanity.com/paper/2305.14251/) (2023)
- [LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression](https://mlsanity.com/paper/2310.06839/) (2023)
- [Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models](https://mlsanity.com/paper/2510.04618/) (2025)
- [ACON: Optimizing Context Compression for Long-horizon LLM Agents](https://mlsanity.com/paper/2510.00615/) (2025)
- [Decontextualization: Making Sentences Stand-Alone](https://mlsanity.com/paper/2102.05169/) (2021)
- [Molecular Facts: Desiderata for Decontextualization in LLM Fact Verification](https://mlsanity.com/paper/2406.20079/) (2024)
- [Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction](https://mlsanity.com/paper/2605.05242/) (2026)
- [Generalization Bias in Large Language Model Summarization of Scientific Research](https://mlsanity.com/paper/2504.00025/) (2025)
- [Fair Abstractive Summarization of Diverse Perspectives](https://mlsanity.com/paper/2311.07884/) (2023)
- [Your AI, Not Your View: The Bias of LLMs in Investment Analysis](https://mlsanity.com/paper/2507.20957/) (2025)
- [Understanding and Improving Information Preservation in Prompt Compression for LLMs](https://mlsanity.com/paper/2503.19114/) (2025)
- [Behavioral Analysis of Information Salience in Large Language Models](https://mlsanity.com/paper/2502.14613/) (2025)
- [LLM as a Broken Telephone: Iterative Generation Distorts Information](https://mlsanity.com/paper/2502.20258/) (2025)
- [Evaluating LLMs in Finance Requires Explicit Bias Consideration](https://mlsanity.com/paper/2602.14233/) (2026)
- [LLMs Corrupt Your Documents When You Delegate](https://mlsanity.com/paper/2604.15597/) (2026)
- [Quantifying Cognitive Bias Induction in LLM-Generated Content](https://mlsanity.com/paper/2507.03194/) (2025)
- [SCOPE: A Generative Approach for LLM Prompt Compression](https://mlsanity.com/paper/2508.15813/) (2025)
- [Document Summarization with Conformal Importance Guarantees](https://mlsanity.com/paper/2509.20461/) (2025)
- [FinTexTS: Financial Text-Paired Time-Series Dataset via Semantic-Based and Multi-Level Pairing](https://mlsanity.com/paper/2603.02702/) (2026)
- [MAD-Fact: A Multi-Agent Debate Framework for Long-Form Factuality Evaluation in LLMs](https://mlsanity.com/paper/2510.22967/) (2025)
- [FinGround: Detecting and Grounding Financial Hallucinations via Atomic Claim Verification](https://mlsanity.com/paper/2604.23588/) (2026)
- [Frame In, Frame Out: Measuring Framing Bias in LLM-Generated News Summaries](https://mlsanity.com/paper/2505.05406/) (2025)
- [From Text to Alpha: Can LLMs Track Evolving Signals in Corporate Disclosures?](https://mlsanity.com/paper/2510.03195/) (2025)
- [Spot the BlindSpots: Systematic Identification and Quantification of Fine-Grained LLM Biases in Contact Center Summaries](https://mlsanity.com/paper/2508.13124/) (2025)
- [Faithful Summarisation under Disagreement via Belief-Level Aggregation](https://mlsanity.com/paper/2601.04889/) (2026)

## Built on by (most cited first, 2 of 2)

- [Compositional Policy Violations: When Step-Level Compliance Fails In Agentic AI Workflows](https://mlsanity.com/paper/2609.18820/) (2026)
- [FinAutoRubric: Expert-Guided Automatic Rubric Generation for Evaluating Financial Research Agents](https://mlsanity.com/paper/2609.35744/) (2026)

## Abstract (arXiv)

Financial decision-makers face more information than they can directly inspect, making context compression necessary. Yet when large language models (LLMs) compress financial source material, they can alter the investment judgment supported by the original source. We frame this problem as information fidelity: compression loses fidelity when it changes the decision induced by the source. In agentic systems, such losses may recur across intermediate steps and amplify throughout the decision process. Across financial filings and earnings-call transcripts, we find that LLM-based compression can produce fluent and factually plausible compressed contexts that nevertheless alter downstream decisions. We analyze two diagnostic patterns associated with fidelity loss: decontextualization, where salient evidence is retained but separated from the caveats and contextual qualifiers needed for correct interpretation, and model dependency, where different compressors expose different views of the same source. We then propose Agentic Context Compression, which generates multiple candidate compressions and audits their disagreements against the original source. Our results suggest that financial compression should be evaluated not only by efficiency or factuality, but also by its ability to preserve decision-relevant context.