The results show that the LLM's capability dissolves with dimension in a way no noise-corrupted classical learner mimics - which explains why LLMs, so capable elsewhere, keep losing to fifty-year-old baselines on tables, while leaving the mechanism of the prediction as an open question.
Abstract
Large language models (LLMs) have become the default tool for a remarkable range of tasks, yet they have had conspicuously little success at one of the most common machine learning workloads: predictive analytics over tabular data. This gap is the founding premise of the fast-growing field of tabular foundation models, but the question of why generic LLMs fail has remained open. We study a frontier LLM in its purest inference regime - a single generation pass over a prompt containing the full training and test data, with no tools, no agentic scaffolding, and no fine-tuning - and systematically evaluate five hypotheses for the failure: (a) an inability to handle noisy or non-linearly-separable data; (b) the linearised CSV format obscuring column structure; (c) the tokenisation of numeric values; (d) the number of test points classified per query; and (e) the dimensionality of the input. Controlled experiments falsify (a)-(d). Dimensionality, in contrast, is decisive: sweeping random linear projections of thirty-one benchmark datasets, the LLM is the only method among nine whose accuracy decreases as dimensionality grows, while every classical baseline stays flat or improves. A behavioural comparison against 252 configured classical models finds that in two dimensions the LLM predicts like a local, distance-based method (up to 91.6% grid agreement), but in higher dimensions no classical model - even when augmented with tuned, dimension-dependent noise - reproduces its predictions. We do not claim to have identified the internal mechanism; our results show, more modestly, that the LLM's capability dissolves with dimension in a way no noise-corrupted classical learner mimics - which explains why LLMs, so capable elsewhere, keep losing to fifty-year-old baselines on tables, while leaving the mechanism of the prediction as an open question.
Tabular Foundation Models (TFMs) increasingly rely on in-context learning, where a model receives labelled examples at inference time and predicts labels for new inputs without updating its weights. Existing TFMs are typically trained on either massive synthetic corpora or very large collections of real datasets. In contrast, we show that surprisingly strong transfer can emerge from self-supervised pre-training on just a single real table. In this setting, we also find that tables tend to be either broadly useful or broadly poor regardless of downstream prediction task, and that the strongest predictor of usefulness is the number of features rather than the number of instances. This leads to a task-centric interpretation of tabular pre-training: the number and the quality of tasks are essential for the pre-training of TFMs. We show that the same task-centric perspective can help corpus design at scale: fine-grained column-level pre-processing consistently improves downstream performance, while no improvements are observed when we filter or deduplicate at the dataset level. Finally, we offer a new perspective for how TFMs generalize: we believe that tabular in-context generalization is largely retrieval-based, and good models are those that learn to identify relevant examples in the provided context and aggregate them well. The mechanics of TFMs have been relatively understudied; our task-centric, retrieval-based perspective offers a new framework to guide future model and corpus design.
Nour Shaheen, Junwei Ma, Alex Labach et al.· 0 citations
Tabular foundation models (TFMs) have moved tabular machine learning from per-dataset training towards amortised in-context inference, fitting a small-to-medium table in a single forward pass without a training run. The 2024–2026 release train, the TabPFN and TabICL lines and challengers such as Mitra, LimiX and Orion, has produced a generation whose architectures, capabilities and limits are documented mainly in preprints, while existing surveys treat these models as a subsection of tabular deep learning or of language-model table understanding. This review is, to our knowledge, the first organised around the current generation. From a corpus of 961 screened records and 98 retained studies, it taxonomises the architectures by pretraining regime, maps the capability space across five axes, isolates the language-model-on-tabular strand for prediction, feature engineering and generation, and summarises openness and deployment. A dedicated critical synthesis then reads the reported capabilities against independent evidence: on the studies reviewed here, tree-based and deep models retain the lead across 142 curated datasets that go beyond the standard independent and identically distributed setting; on 112 datasets, the models attain the highest accuracy but weaker conditional coverage than gradient-boosted trees; and robustness under feature shift, fairness and generation quality remain open. Amortised in-context prediction is thus a working paradigm whose independent evidence has yet to match its benchmark claims.
Sergei O. Kurashkin, V. Tynchenko, Alexey S. Borodulin et al.· Machine Learning and Knowled...· 0 citations
Predicting missing cell values in tabular data is a fundamental problem in data cleaning. While state-of-the-art reasoning models show great promise in predicting missing values in tables, by reasoning holistically across rows and columns, they are costly to deploy at scale and tend to be overconfident, often generating hallucinated or false-positive predictions. In this paper, we observe that achieving high-precision missing-value prediction in tables requires a distinct combination of three capabilities: (1) world knowledge, (2) text-based reasoning, and (3) code-based reasoning. We systematically explore design choices for combining these capabilities, and propose an Auto-Fill approach that post-trains three specialist small language models (SLMs), each optimized for one capability. We develop a calibrated ensemble mechanism that either dynamically selects the most confident specialist or abstains, ensuring high accuracy. Extensive experiments on 11 benchmarks with 2200 real tables drawn from diverse domains show that Auto-Fill achieves superior accuracy compared to state-of-the-art reasoning models (e.g., o3-pro, Gemini 3 Pro, and DeepSeek R1), while operating at a fraction (less than 1%) of the cost of these frontier models. Our results highlight the effectiveness of specialization and calibrated abstention in the important domain of tabular data. Auto-Fill is publicly available at https://github.com/lyrain2001/auto-fill.
Yurong Liu, Yeye He, Haoyu Dong et al.· 0 citations
PredicateLongBench is proposed, a benchmark that stress-tests long-context reasoning by asking models to identify the longest contiguous subsequence of words in a long input that satisfies given predicates/constraints drawn from a broader predicate class.
We study the problem of sequentially evaluating a new large language model (LLM) on a fixed question set using historical performance data from prior LLMs. Our goal is to construct a confidence sequence (CS) for the model's capability on this question set and to design active querying rules that shrink the CS width as quickly as possible. For CS construction, we invert a family of test supermartingales and focus on two representative approaches: a reverse information projection (RIPr)-based approach and a testing-by-betting-based approach. We first study these approaches under an oracle setting, and demonstrate the oracle optimality of the RIPr-based construction. We then propose a growth-oriented querying rule that aims to maximize the worst-case one-step expected log-increment over the endpoints of the current CS. In practice, we build these test supermartingales and the querying rule on predictions of question-level correctness learned from historical data. We then analyze the shrinkage behavior of the resulting CSs and identify two key factors that slow the shrinkage rate of CSs: accumulated prediction mismatch and the spikiness of the querying distribution. Finally, motivated by this analysis, we propose several mixture querying rules that combine growth-oriented querying, prediction refinement, and uniform exploration, trying to mitigate the effects that slow the shrinkage rate. We provide experiments comparing different querying rules for the RIPr-based and testing-by-betting-based CSs across several synthetic testing datasets. Interestingly, we observe that the simplest querying rule, uniform sampling, can sometimes outperform more adaptive querying rules for both methods.
Large Language Models (LLMs) have emerged as powerful tools for processing the heterogeneous information environments of modern financial markets. This paper presents a systematic, comparative evaluation of five prominent LLMs: GPT-4 Turbo, Claude 3 Opus, Gemini 1.5 Pro, Llama 3 70B, and the domain-specialized FinGPT, with respect to their capacity for technical market analysis. The evaluation spans four structured tasks: candlestick pattern recognition from OHLCV data, directional signal generation (BUY/SELL/HOLD), backtesting of signal quality through a simulated execution pipeline, and financial report comprehension. Our experimental framework employs rigorous quantitative metrics, including Sharpe ratio, maximum drawdown, Sortino ratio, information coefficient, F1-score, and BLEU score. Findings from simulated backtesting indicate that GPT-4 Turbo achieves the highest annualized return and Sharpe ratio among general-purpose models, while FinGPT demonstrates competitive risk-adjusted performance due to domain-specific fine-tuning. Both models outperform a passive S&P 500 benchmark under the tested conditions. The study identifies persistent failure modes across all evaluated models, including numerical hallucination, context-window limitations, and inconsistent performance in sideways market regimes. We conclude that while LLMs hold genuine promise within AI trading systems, robust deployment requires careful task decomposition, rigorous backtesting protocols, and domain-aware fine-tuning strategies.