Detecting whether corporate sustainability disclosures truly satisfy reporting standards is essential for greenwashing-aware ESG auditing, yet existing resources rarely evaluate whether disclosures sufficiently satisfy framework metrics. We introduce MM-COMPLIANCE, a multilingual multimodal dataset aligned with SASB metrics across six languages (English, French, Japanese, Chinese, Thai, and Korean), with page-level evidence and a key completeness annotation that separates fully compliant disclosures from partial mentions. We define two tasks reflecting practical auditing workflows: Page-level Compliance Judgment and Metric-to-Page Evidence Retrieval. Our evaluated baselines primarily comprising open-source architectures, supplemented by a commercial VLM as an upper-bound reference, show moderate performance for page understanding and evidence localization, but consistently low performance on partial compliance verification and large variation across languages and industries. MM-COMPLIANCE provides a benchmark for developing framework-aware models to support rigorous ESG compliance auditing and long-term commitment tracing.

Demand for personalized financial advice is growing, yet current LLM-based advisors often fail to provide consistent and specialized guidance. Simple persona prompts rarely specify how a financial advisor should reason and often drift toward generic recommendations. We propose Fund2Persona, a framework that builds financial-advisor personas from real-world fund disclosures and refines them through an actor–scorer–patcher loop. We test whether the resulting personas can predict held-out portfolio changes and produce commentary consistent with fund managers’ own explanations. They outperform generic baselines on both tasks. We further study two downstream diagnostics: market-scenario generation, where persona retrieval broadens plausible investment views, and multi-turn investor–advisor conversations, where matched personas give more specific and useful advice than a generic advisor. These results suggest that real fund data can bring manager-specific investment expertise to LLM advisors rather than merely changing an LLM’s surface style.

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.