Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents
This paper studies behavioral alignment and representation dynamics of LLM agents in financial environments using TradeArena. It identifies measurable pre-failure signatures like planning embedding drift and effective-rank contraction. Structured risk feedback can serve as an external alignment signal but is not a universal performance enhancer. A 51-stock experiment reveals a correlation blind spot where LLM rationales justify concentrated exposure to coupled assets.
Article intelligence
Key points
- LLM agents exhibit measurable pre-failure signatures including planning embedding drift and effective-rank contraction.
- Structured risk feedback acts as an external alignment signal but varies in effectiveness across models.
- A 51-stock experiment uncovers a correlation blind spot: LLM rationales often justify concentrated exposure to correlated assets that the risk layer repeatedly clips.
Why it matters
This matters because LLM agents exhibit measurable pre-failure signatures including planning embedding drift and effective-rank contraction.
Technical impact
May affect model selection, inference cost, product capability, and evaluation benchmarks.
[2605.28850] Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents
[Submitted on 16 May 2026]
Title:Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents
View a PDF of the paper titled Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents, by Weicheng Xue
View PDF HTML (experimental)
Abstract:We study behavioral alignment and representation dynamics of large language model (LLM) agents in financial decision environments. Using TradeArena, an auditable trading-agent testbed with risk reports, execution simulation, memory, and replayable trajectories, we analyze how rationales, positions, and interventions evolve under market stress. We find measurable pre-failure signatures: planning embeddings drift from normal-state centroids, fused plan-risk representations separate normal from pre-drawdown states, and manifold diagnostics show effective-rank contraction before failures. To address small-sample and embedding-choice concerns, we use 80 rolling failure anchors across eight LLM trajectories and show that contraction persists across hash, LSA, Transformer, and white-box hidden-state probes. Stress tests with CoT-free target weights, lexical controls, OHLCV noise, and false-audit reports indicate that rationale-level contraction can vanish without rationales, while intent-space contraction may remain; lexical diversity does not collapse; and fused signatures remain informative under noise. We also find that structured risk feedback can act as an external alignment signal without fine-tuning, but not as a universal performance enhancer: true audit feedback improves calibration for some models, return and drawdown for others, and reveals cases where hidden or placebo feedback has higher short-horizon return but weaker alignment diagnostics. Finally, a 51-stock intraday experiment reveals a correlation blind spot: LLM rationales often justify concentrated exposure to coupled assets that the risk layer repeatedly clips, with a rolling Markowitz baseline as a covariance reference. These results support a research claim rather than a profitability claim: auditable risk feedback and representation trajectories reveal when LLM financial reasoning is aligning, drifting, or failing.
Subjects:
Machine Learning (cs.LG); Computational Finance (q-fin.CP)
Cite as: arXiv:2605.28850 [cs.LG]
(or arXiv:2605.28850v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2605.28850
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Weicheng Xue [view email] [v1] Sat, 16 May 2026 16:53:34 UTC (38 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Representation Signatures and Risk-Feedback Alignment in LLM Trading Agents, by Weicheng Xue
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.LG
new | recent | 2026-05
Change to browse by:
cs q-fin q-fin.CP
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
IArxiv recommender toggle
IArxiv Recommender (What is IArxiv?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)