ML-Quant
Papers
6,576 papers featured in the letter since May 2023, each with our one-sentence summary and, where Semantic Scholar knows it, what became of it.
- arXiv
- 2,131
- SSRN
- 2,801
- RePEc
- 808
- Machine learning
- 836
By venue
arXiv
Quantitative-finance and ML-for-finance preprints from arXiv.
- Agentic AI Systems and Financial Stability, From Model Risk to Systemic Risk
- FactorBench: A Portfolio-Aware Benchmark for Automated Factor Mining
- Towards Financial World Modeling
2,131 paperssince 24 May 2023
SSRN
Working papers in finance and economics from SSRN.
- The Evolution of Quantitative Equity Investing: Half a Century of Alpha Research and Institutional Adaptation
- Does Autonomous Quant Research Improve? Leakage, Evidence Scarcity and Forward Tests in an LLM Factor-Discovery Loop
- The Three Faces of Fed Surprise: Rate-Path, Information and Tone Shocks in the Equity Cross-section
2,801 paperssince 24 May 2023
RePEc
Economics working papers from RePEc's NEP field reports.
- Risk Appetite and Monetary Transmission
- Estimating Yield Impacts of Treasury Demand and Supply Changes
- Collateral Advantage: Exchange Rates, Capital Flows, and Global Cycles
808 paperssince 24 May 2023
Machine learning
The general machine-learning papers the letter carried in 2023-25.
- Fairness in Survival Analysis: A Novel Conditional Mutual Information Augmentation Approach
- A comparison of translation performance between DeepL and Supertext
- Revisiting Expected Possession Value in Football: Introducing a Benchmark, U-Net Architecture, and Reward and Risk for Passes
836 paperssince 16 Oct 2023
Most cited
Featured papers with the most citations today (Semantic Scholar).
- 7 Feb 20249,718cites
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Advancing Math Reasoning in Language Models: DeepSeekMath7B is a new language model that uses web data and Group Relative Policy Optimization for advanced mathematical reasoning, scoring high on the MATH benchmark.
Machine learningLLMs & Text
- 5 Jun 20249,474cites
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
Sequence Modeling: Mamba, a neural network architecture that doesn't use attention or MLP blocks, provides faster inference and better performance in language, audio, and genomics than Transformers.
Machine learningML & AI Methods
- 16 Oct 20234,011cites
Mistral 7B
Superior Language Model: Mistral 7B v0.1 is a language model with 7 billion parameters that excels in reasoning, mathematics, and code generation, and has a version specifically designed to follow instructions.
Machine learningLLMs & Text
- 9 Oct 20242,343cites
MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark
MMLU-Pro, an improved dataset, expands the Massive Multitask Language Understanding benchmark by adding tougher questions and more choices, serving as a better benchmark to monitor progress in the field.
Machine learningOtherIn Neural Information Processing Systems
- 20 Jun 20242,329cites
Depth Anything V2
Depth Anything V2 is a new model for monocular depth estimation, using synthetic and large-scale pseudo-labeled real images for faster, more accurate results and setting a new evaluation benchmark.
Machine learningOtherIn Neural Information Processing Systems
- 7 Aug 20242,266cites
Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters
The research investigates enhancing Large Language Models' (LLMs) performance using more test-time computation, suggesting a compute-optimal scaling strategy based on prompt difficulty.
Machine learningLLMs & Text
- 5 Jun 20242,044cites
Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
The research identifies a link between state-space models and Transformers in deep learning, leading to the creation of a faster language modeling architecture, Mamba-2.
Machine learningML & AI MethodsIn International Conference on Machine Learning
- 22 May 20242,026cites
Octo: An Open-Source Generalist Robot Policy
Octo is a large transformer-based policy for robotic manipulation, trained on a vast dataset, that can be instructed via language or images and adapted to new domains.
Machine learningML & AI MethodsIn Robotics: Science and Systems Conference
- 24 Jul 20241,914cites
AWQ: Activation-aware Weight Quantization for On-Device LLM Compression and Acceleration
The study suggests Activation-aware Weight Quantization (AWQ), a hardware-friendly method for quantizing large language models that reduces error and improves performance on various benchmarks.
Machine learningLLMs & TextIn GetMobile: Mobile Computing and Communications
- 12 Dec 20241,802cites
Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
The paper presents InternVL 2.5, a sophisticated multimodal large language model that performs well on various benchmarks, exceeding 70% on the MMMU benchmark.
Machine learningLLMs & Text
- 25 Sep 20241,628cites
Qwen2.5-Coder Technical Report
The report unveils the Qwen2.5-Coder series, an improvement from its predecessor, showcasing remarkable code generation abilities and achieving top-tier performance in various code-related tasks.
Machine learningOtherFeatured 2×
- 16 Oct 20231,558cites
MemGPT: Towards LLMs as Operating Systems
Extended Context in LLMs: MemGPT is a system that manages different memory levels, providing extended context within large language models' limited context windows, enhancing document analysis and multi-session chat performance.
Machine learningLLMs & TextFeatured 2×
- 5 Feb 20251,521cites
s1: Simple test-time scaling
The research presents a method called budget forcing, which uses a small dataset to achieve test-time scaling and improved reasoning performance in language modeling, particularly in competition math questions.
Machine learningLLMs & TextIn Conference on Empirical Methods in Natural Language Processing
- 22 May 20241,507cites
DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
MoE Language Model: DeepSeek-V2, a language model with 236B parameters, offers enhanced performance and cost efficiency compared to its predecessor, ranking high among open-source models.
Machine learningLLMs & Text
- 24 Apr 20241,485cites
Mastering Diverse Domains through World Models
Algorithm Mastery: DreamerV3, a universal algorithm, excels in over 150 varied tasks, including diamond collection in Minecraft without human input, expanding the scope of reinforcement learning.
Machine learningML & AI Methods
- 16 Oct 20231,302cites
SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models
Hallucination Detection for LLMs: The paper presents SelfCheckGPT, a new approach for fact-checking black-box model responses without an external database, proving its superior ability to detect and rank factual and non-factual sentences.
Machine learningLLMs & TextIn Conference on Empirical Methods in Natural Language ProcessingFeatured 2×
- 12 Jun 20241,209cites
Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction
The article discusses Visual AutoRegressive modeling (VAR), a new image learning method that outperforms diffusion transformers in terms of speed, image quality, and scalability.
Machine learningML & AI MethodsIn Neural Information Processing Systems
- 10 Jul 20241,206cites
SimPO: Simple Preference Optimization with a Reference-Free Reward
Simple Preference Optimization: SimPO improves reinforcement learning from human feedback by using the average log probability of a sequence as the implicit reward, enhancing training stability and computational efficiency.
Machine learningML & AI MethodsIn Neural Information Processing Systems
- 8 May 20241,012cites
A Simple and Effective Pruning Approach for Large Language Models
Wanda, a new method, efficiently prunes weights in Large Language Models without retraining, offering a more efficient approach to inducing sparsity in pretrained models.
Machine learningLLMs & TextIn International Conference on Learning Representations
- 3 Jan 2024993cites
PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis
PIXART-$\alpha$, a Transformer-based text-to-image model, generates high-quality images at a low cost, reducing CO2 emissions and offering a cost-effective solution for the AIGC community.
Machine learningML & AI MethodsIn International Conference on Learning Representations
- 5 Feb 2025975cites
TÜLU 3: Pushing Frontiers in Open Language Model Post-Training
Open Language Model Post-Training: The Tulu 3 model, a top-tier post-trained language model, is introduced, outperforming other models and providing a detailed guide for its use and adaptation.
Machine learningLLMs & Text
- 12 Jun 2024947cites
Simple and Effective Masked Diffusion Language Models
The performance of diffusion models in language modeling has been enhanced by using an effective training recipe and a simplified objective, setting a new standard among diffusion models.
Machine learningLLMs & TextIn Neural Information Processing Systems
- 9 Jan 2024934cites
TinyLlama: An Open-Source Small Language Model
Small Open-Source Language Model: The article presents TinyLlama, a compact 1.1B language model that performs remarkably well in various tasks despite its small size, having been pretrained on around 1 trillion tokens.
Machine learningLLMs & Text
- 12 Dec 2024811cites
Training Large Language Models to Reason in a Continuous Latent Space
The article presents Coconut, a new approach that uses the last hidden state of large language models for reasoning in an unrestricted latent space, proving its effectiveness in enhancing the LLM on multiple reasoning tasks.
Machine learningLLMs & Text
- 8 May 2024765cites
Fourier Neural Operator with Learned Deformations for PDEs on General Geometries
The study introduces geo-FNO, a new framework for solving partial differential equations on any geometry, proving to be faster and more accurate than standard and machine learning-based solvers.
Machine learningML & AI MethodsIn J. Mach. Learn. Res.
Latest featured
- 9 Oct 20263fanfare
Firm's refinancing intensity, the cost of capital, and real investment
A global study of 357,630 firm-years shows firms with more short-term debt cut capital expenditure by 2.156 percentage points, mainly due to refinancing availability constraints.
- 9 Oct 20263fanfare
Exchange Rate Regimes and Term-Premium Spillovers from U.S. Monetary Policy
U.S. monetary tightening raises term premia sharply in pegged economies but not floating ones; a DSGE model shows pegs force adjustment through domestic rates, tightening leverage constraints and amplifying spillovers.
- 9 Oct 20263fanfare
Expectations and the Transmission of Monetary Policy
Optimal rate-setting creates time-inconsistency: market expectations become too sensitive to policy surprises, amplifying real effects of central bank forecast errors; less aggressive rules cut output volatility.
- 9 Oct 20263fanfare
Common Banks, Supply Chains, and the Transmission of Real Shocks
Using Portuguese wildfire data, the research finds firms whose banks also lend to suppliers experience 10 percentage points smaller credit declines and face attenuated real losses.
- 9 Oct 20264fanfare
Monetary-Policy Risk and Equilibrium Asset Prices
Asset prices respond to monetary policy uncertainty around FOMC announcements; a New Keynesian model shows policy risk accounts for much of equity and inflation-bond risk premia and weakens activity when elevated.
- 9 Oct 20264fanfare
Volatility Targeting, Variance Drain, and the Financialization Effect: A Century of Evidence
Analyzing 100 years of daily data, the study finds volatility scaling harmed returns before 1985 but improved them after, a shift attributed to derivatives markets and deregulation, with walk-forward Sharpe ratio of 1.01.
- 9 Oct 20263fanfare
Delta-Guided Soft Actor-Critic for Derivative Hedgingunder Model Misspecification and Transaction Costs
A reinforcement learning method embedding Black-Scholes delta guidance cuts hedging error by 19-38% across simulated and real options data with robust statistical significance.
- 9 Oct 20263fanfare
Geoeconomic Fear and Sovereign Spreads
The study constructs a geoeconomic fear index from news articles and finds that higher fear predicts wider emerging-market sovereign spreads, especially for speculative-grade issuers.
SSRNLLMs & Text
- 9 Oct 20263fanfare
Predicting Crypto Contagion: Machine Learning Forecasts of Return Spillovers
The study forecasts time-varying return spillovers among cryptocurrencies using machine learning, finding monetary policy and funding rates drive connectedness and classifying events that shift spillover composition into contagion types.
SSRNCrypto & DeFi
- 9 Oct 20263fanfare
Artificial Intelligence Exposure and Financial Stability: Cross-Industry Common Risk and Connectedness in U.S. Equity Markets
The study measures AI exposure across 284 U.S. firms and finds that an AI-factor explains only 18% variance alongside conventional factors, suggesting AI is a distinct cross-industry risk lens separate from traditional spillovers.
- 9 Oct 20262fanfare
Large Language Models and Information Frictions in Corporate Bonds
Shows that large-language-model scores of default risk from earnings calls predict bond rating migration one year ahead where analysts disagreed, capturing information not yet priced in spreads.
SSRNLLMs & Text
- 9 Oct 20263fanfare
GFlowAlpha: A Quality-Control-Aware Neuro-Symbolic Framework for White-Box Alpha Factor Mining
GFlowAlpha uses neuro-symbolic search and internalizes econometric quality control as online rewards to mine white-box factors, delivering mean out-of-sample daily ICIR of 0.485 with Fama-MacBeth t-statistics of 4.99 on Chinese stocks.
SSRNML & AI Methods
- 9 Oct 20263fanfare
A hybrid graph neural network for realized volatility forecasting: Multi-relation attention and dual-path fusion
A graph neural network combining temporal and cross-asset volatility patterns reduces mean squared error and quasi-likelihood loss by approximately 9.9% and 3.4% on Dow stocks.
SSRNML & AI Methods
- 9 Oct 20264fanfare
Firm-level AI Exposure: Measurement, Determinants, and Pricing Implications
A natural-language analysis of 13,757 firms finds AI exposure surged after ChatGPT and predicts 1.37 percent higher annual returns, though the effect is concentrated after the launch and driven by firm-specific factors.
SSRNML & AI Methods
- 9 Oct 20263fanfare
The Price of Conviction
The study shows that high-conviction beliefs exert outsized influence on prices, with a long-short portfolio exploiting conviction-weighted disagreement earning 13.2 percent annualized six-factor alpha.
- 9 Oct 20263fanfare
Global Geopolitical Skewness Risk and Bond Risk Premia
A measure of cross-country geopolitical risk skewness predicts U.S. Treasury risk premia out-of-sample and forecasts weaker activity, adding incremental value beyond yield curves and standard macroeconomic predictors.
- 9 Oct 20263fanfare
The Credit Spread at Long Horizons
The study shows that Baa-Aaa spreads at 27-year maturity test pricing kernels better than shorter maturities, with a structural refinancing model matching observed spreads while prior kernels overpredict by 27 to 68 basis points.
- 9 Oct 20264fanfare
Supplying Safety to an Aging World: Demographics, Debt, and Risk Premia
A two-country model attributes 251 basis points of U.S. equity premium gains and 71 basis points of global rate decline since 1980 to demographic aging.
- 9 Oct 20263fanfare
Agents Propose, Code Decides: An Autonomous Research Loop Framework to Build an Equity Alpha Model
An autonomous research loop in which agents propose factor candidates and code verifies them builds multi-factor alpha models, screening 212 predictors and admitting 14 that raise baseline information coefficient out-of-sample.
SSRNML & AI Methods
- 9 Oct 20263fanfare
Fundamentals in Cryptocurrency Valuation: Evidence from DeFi Markets
Revenue surprises on 95 DeFi protocols predict abnormal token returns, with one-standard-deviation surprises associated with 28.8 basis points higher returns over three days when tokens have direct revenue claims.
SSRNCrypto & DeFi
- 9 Oct 20264fanfare
The Evolution of Quantitative Equity Investing: Half a Century of Alpha Research and Institutional Adaptation
Surveys fifty years of quantitative investing as a history of changing implementation boundaries from diversification through machine learning, emphasizing alpha decay and the gap between research and production systems.
- 9 Oct 20263fanfare
Reassessing Bitcoin’s Role under ETF-Driven Institutionalization: A Time-Frequency Connectedness Analysis
The research shows that after U.S. Bitcoin ETF approval in January 2024, return and volatility linkages between Bitcoin and equities strengthened substantially, with volatility spillovers becoming markedly asymmetric from stocks to crypto.
SSRNCrypto & DeFi
- 9 Oct 20263fanfare
Cross-Sectional Corporate Bond Returns
Constructs a characteristics-based systematic risk measure using rate jumps that predicts out-of-sample corporate bond returns, with high-minus-low decile portfolio earning 9.12% annually and 7.80% alpha.
- 9 Oct 20264fanfare
A Multi-Strategy Quantitative Portfolio
Eleven systematic strategies across equities, futures, and currencies achieve a 1.91 Sharpe ratio net of costs from January 2012 to June 2026, with 23.6% compound annual return.
- 9 Oct 20264fanfare
The Three Faces of Fed Surprise: Rate-Path, Information and Tone Shocks in the Equity Cross-section
The study decomposes Federal Reserve announcements into three orthogonal shocks and traces each through the equity cross-section, finding rate-path surprises dominate while standard balance-sheet characteristics carry no additional power beyond beta and size.
- 9 Oct 20263fanfare
Deflate by Bits, Not Trials: Why Query-Counted Deflated Sharpe Ratios Fail Under Adaptive Search, and a Sealed-Holdout Remedy
The research proves that query-counted Deflated Sharpe Ratios certify noise in adaptive research and proposes sealed-holdout verification, where validity depends on bits revealed rather than strategy count.
SSRNML & AI Methods
- 9 Oct 20263fanfare
Heterogeneous Estimation of the Causal Effect of Debt Restructuring Interventions: A Causal and Explainable Uplift Modeling Framework for the Economic Optimization of Credit Risk Portfolios
A causal framework compares six estimators of treatment effect to optimally assign debt restructuring interventions, revealing that targeting borrowers by predicted effect generates substantially positive value where blanket treatment fails.
- 9 Oct 20264fanfare
Does Autonomous Quant Research Improve? Leakage, Evidence Scarcity and Forward Tests in an LLM Factor-Discovery Loop
An autonomous agent evolved 940 factors over 17 days and finds that reusing backtest data inflates edge by a quarter to a third and in-sample improvement predicts worse performance.
SSRNML & AI Methods
- 9 Oct 20263fanfare
What Survives Honest Tes1ng? A Pre-registered Test of Eleven Trading Signals
Applies a strict pre-registered protocol to eleven trading signals across equity and futures markets, finding only trend survives development and hold-out periods with borderline significance after multiple-testing adjustment.
- 9 Oct 20263fanfare
The Toll-Fenced Market: Documented Anomalies in Indian Equities Replicate but do Not Survive Retail Implementation Costs
Well-documented Indian equity anomalies replicate strongly in-sample but all fail to survive 12 to 32 basis point retail transaction costs; required directional accuracy reaches 72% at hourly horizons.
- 9 Oct 20262fanfare
Housing Network Connectedness and Policy Spillovers: Evidence from a Time-Varying Parameter VAR Approach
A time-varying VAR network reveals supply-inelastic coastal markets absorb largest regulatory shocks and spillovers persist to monetary conditions and inflation expectations.
- 9 Oct 20263fanfare
From Tone to Trajectory: Continuous Sentiment and the Shape of Monetary Policy Communication (Martin Feldkircher, Márton Kardos, Kristoffer Laigaard Nielbo)
The sequencing and emphasis of sentiment within central bank statements predict rate decisions and inflation expectations beyond average tone alone.
- 9 Oct 20262fanfare
Disappearing Dollar Convenience?
Dollar safe asset convenience yields remain stable domestically since 2019 but decline globally relative to other currencies, with repo markets leading.
- 9 Oct 20263fanfare
A Model of Global Firms and Capital Markets
A quantified general-equilibrium model of 23,000 firms across 48 countries shows financial and trade liberalization concentrate activity in largest firms with larger gains for emerging economies.
- 9 Oct 20263fanfare
A Currency Premium Puzzle
The paper proves that asset pricing models solving equity and risk-free rate puzzles fail to generate observed interest-rate differentials between risky and safe currencies.
- 9 Oct 20263fanfare
The Equilibrium Impact of Credit Frictions: Evidence from Default Risk Using Firm-Level Data
UK firm-level data shows relaxing credit frictions raises output by 25% and wages by 23%, with most gains from capital accumulation rather than reallocation.
- 9 Oct 20263fanfare
Beliefs That Predict Returns and Beliefs That Attract Flows: Policy Insights and Sentiment Catering in Mutual Funds
The study shows that mutual fund managers' policy beliefs predict market returns and explain positive alphas, while sentiment beliefs attract flows but lack predictive power.
- 9 Oct 20263fanfare
Volatility Disagreement in the Options Market
Cross-sectional dispersion in volatility forecasts predicts delta-hedged straddle losses of 5.14% per month, consistent with mispricing rather than risk compensation.
- 9 Oct 20262fanfare
Optimal Currency Strategies Under Deviations From Interest Parity
Extending Campbell et al., the analysis shows empirically measured deviations from interest parity reverse optimal currency demands for emerging-market investors.
- 9 Oct 20263fanfare
Drivers of Exchange Rates in EMDEs: Implications for Foreign Exchange Intervention
A framework using deviations from interest-rate parity identifies when exchange-rate movements stem from financial shocks rather than fundamentals, aiding intervention decisions.