Liquidity Management Attacks on Lending Markets
Bibliographic record
Abstract
Decentralized Finance (DeFi) continues to open up promising opportunities for a broad spectrum of users, with lending pools emerging as a cornerstone of its applications. While prominent platforms like Compound and Aave maintain a large share of the funds in lending pools, numerous other smaller pools also exist. Many of these smaller entities draw heavily from the design principles of their larger counterparts due to the complex nature of lending pool design. This paper asserts that the design approaches that serve larger pools effectively may not necessarily be the most beneficial for smaller lending pools. We identify and elaborate on two liquidity management attacks, which can allow well-funded attackers to exploit specific circumstances within lending pools for personal gain. Although large lending pools, due to their vast and diverse liquidity and high user engagement, are generally less vulnerable to these attacks, smaller lending protocols may need to employ specialized defensive strategies, particularly during periods of low liquidity. We also show that beyond the six leading lending protocols, there exists a market value exceeding $1.75 billion. This considerable sum is dispersed among over 200 liquidity pools, posing a potentially attractive target for bad actors. Furthermore, we evaluate existing designs of lending pools and suggest a novel architecture that distinctly separates the liquidity and logic layers. This unique setup gives smaller pools the adaptability they need to link with larger, well-established pools. Despite encountering certain constraints, these emerging pools can leverage the considerable liquidity from larger pools until they generate sufficient funds to form their own standalone liquidity pools. This design cultivates a setting where multiple lending pools can integrate their liquidity components, thus encouraging a more diverse and robust liquidity environment.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.001 | 0.005 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.006 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".