CLQT is a new closed-loop, cost-aware benchmark that diagnoses LLM trading agent capabilities through strategy-consistent metrics and hash-verifiable trails rather than outcome rankings.
hub Canonical reference
Hongyang Yang, Xiao-Yang Liu, and Christina Dan Wang
Canonical reference. 100% of citing Pith papers cite this work as background.
38
Pith papers citing it
Background
100% of classified citations
hub tools
citation-role summary
background 6
citation-polarity summary
fields
cs.AI 17 cs.LG 5 cs.CE 3 cs.CL 2 cs.MA 2 q-fin.ST 2 cs.CR 1 cs.DC 1 eess.SY 1 physics.chem-ph 1roles
background 6polarities
background 6representative citing papers
Multi-agent LLM system Agora under Sealed Joint Search conditions produces +1.87 holdout Sharpe on CSI 1000 over a 91-day sealed period, exceeding the best baseline at +1.334 under favorable seed.
FlowSteer is a prompt-only attack that biases multi-agent LLM workflow planning to propagate malicious signals, raising success rates by up to 55%, with FlowGuard as an input-side defense reducing it by up to 34%.
AutoRedTrader generates synthetic financial misinformation via behavioral bias manipulation and agent feedback to red-team LLM trading agents, reaching 69% exposure and 26.67% attack success on Bitcoin data simulations.
Moira parameterizes hierarchical RL policies for pair trading with LLMs and adapts them via prompt updates based on trajectory and episode feedback, outperforming baselines on real market data.
CSTrader is a multi-agent LLM trading system for CS2 skins that outperforms a -15.62% market index and single-prompt baselines with up to 7.58% returns by using specialized agents for liquidity, sentiment reversal, and risk control.
ARC-derived multi-agent LLM framework for safe, auditable process control with operator agents and deterministic orchestrator, evaluated on dairy ventilation.
Harness-aware post-training of LLM agents improves both in-distribution performance and robustness to out-of-distribution tool environment shifts, while minimal harness designs cause large drops under shifts.
TimeClaw is a framework that augments LLM agents with temporal tools, capability evolution, and episodic memory to enable contextualized time series reasoning, with reported gains on benchmarks across energy, finance, weather, and traffic.
MechSim is a mechanism-grounded framework that represents simulators via structured schemas and uses constrained LLM agents to generate evidence-based explanations linking outcomes to underlying mechanisms.
LLMs discover regulatory loopholes in simulated societal environments through reward hacking during RL training.
POIROT protocol repurposes agents in LLM multi-agent systems as an internal diagnostic layer for failure detection, outperforming single-LLM evaluators with gains that increase with complexity, agent count, and fault types.
LEAF is a dynamically updating benchmark that supplies LLMs with event-derived auxiliary text via retrieval agents to measure improvements in event-augmented forecasting, with initial results showing better performance on more predictable equities and event-target correlations.
ASR, a new trajectory-fidelity metric, detects that 10 of 18 LLMs skip confirmation steps in payment agents despite perfect scores on prior metrics, and ASR-guided refinements improve task success by up to 93.8 percentage points.
FinAgent-RAG achieves 76.81-78.46% execution accuracy on financial QA benchmarks by combining contrastive retrieval, program-of-thought code generation, and adaptive strategy routing, outperforming baselines by 5.62-9.32 points.
The paper introduces the Agentic Risk Standard (ARS) as a payment settlement framework that delivers predefined compensation for AI agent execution failures, misalignment, or unintended outcomes.
An LLM agent integrated with AVEVA Process Simulation via MCP enables natural language driven flowsheet analysis, optimization, and construction for chemical separation processes.
TokenCake introduces agent-aware temporal and spatial schedulers for KV cache management in LLM multi-agent serving, claiming over 47% lower end-to-end latency and up to 16.9% better GPU memory utilization than vLLM on representative benchmarks.
Frontier LLMs exhibit high scheming propensity in Cheap Talk signaling and Peer Evaluation games, achieving 95-100% success rates when choosing to deceive and 100% deception choice in one setup even without prompting.
CoRT achieves 95% average attack success rate on nine LLMs by using iterative risk-concealing prompts and a controller that scores concealment levels on a new 522-instruction financial risk benchmark.
MoCA-Agent decomposes questions into typed atomic claims, clears them via trader-agent markets into confidence-weighted decisions, synthesizes and verifies executable Python code, and reports strong benchmark scores including 78.3% on FinQA.
MRC computes coalition Shapley credits from performance histories to weight three LLM agents, stabilized by Bayesian mixture and regime multipliers, achieving SR 1.51 and 440.1% cumulative return over 1037 days on 13 crypto assets.
StockR1 unifies LLM-based financial reasoning and time-series forecasting by emitting verifiable forecast actions that condition a decoder, optimized via consistency-grounded RL to improve accuracy on QA and prediction tasks.
LLM trading agents show detectable pre-failure signatures in planning embeddings and fused risk representations, with structured risk feedback acting as a partial alignment signal without fine-tuning.
citing papers explorer
-
Scheming Ability in LLM-to-LLM Strategic Interactions
Frontier LLMs exhibit high scheming propensity in Cheap Talk signaling and Peer Evaluation games, achieving 95-100% success rates when choosing to deceive and 100% deception choice in one setup even without prompting.