---
title: 'LLMs as Predictors: Methods & Applications'
url: https://www.emergentmind.com/topics/llms-as-predictors
type: topic
---

# LLMs as Predictors: Methods & Applications

Large language models (LLMs) as predictors refer to the direct use of pretrained LLMs—models originally trained for language understanding and generation—as engines for making predictions in supervised, semi-supervised, or unsupervised tasks across domains such as graphs, time series, tabular data, mobility, and even social or behavioral science. Distinct from traditional architectures purpose-built for specific data types (e.g., graph neural networks for graphs, or LSTMs for time series), LLMs-as-predictors rely on prompt-driven reformulation of prediction tasks into text-based problems, leveraging the models' world knowledge, semantic reasoning, and flexible input handling. This paradigm encompasses zero-shot, few-shot, and prompt-tuned approaches, as well as frameworks that combine LLM predictive outputs with conventional machine learning models or pipelines.

## 1. Key Principles and Pipeline Design

The foundational principle behind LLMs-as-predictors is the “translation” of prediction problems into natural language or structured prompt formats interpretable by the LLM. For node classification on graphs, for example, each node is encoded as a prompt using its textual attributes, optionally with neighborhood context rendered as text (e.g., a summary of 2-hop neighbors), and the LLM is tasked with directly generating the node label [2307.03393]. Similar pipelines are constructed for time series (where sequences are decomposed and represented as text) [2506.02389], mobility forecasting (listing historical “stay” events in textual form) [2308.15197, 2405.20962], and tabular data (prompting with structured rows in a CSV/JSON or Markdown-like format) [2508.17391, 2508.19563].

Prompting strategies may include:
- **Zero-shot:** Task is specified as a question involving only the target sample
- **Few-shot or in-context learning:** A small set of labeled examples (input–output pairs) are provided prior to the query sample.
- **Chain-of-thought prompting:** The LLM is instructed to reason step by step before issuing a prediction.

The output is typically unstructured text and requires postprocessing—e.g., parsing text to label names via string matching or edit distance—as LLMs do not natively constrain outputs to a discrete label set [2307.03393].

## 2. Applications Across Domains

LLMs-as-predictors have been studied in diverse settings:

| Domain       | Task Type                     | Data/Pipeline Example                                     |
|--------------|-------------------------------|----------------------------------------------------------|
| Graphs       | Node classification           | Node text + neighbor summary → LLM → label [2307.03393]  |
| Mobility     | Next-location prediction      | Historical/context stays → prompt → ranked locations [2308.15197, 2405.20962] |
| Social sci.  | Social feature prediction     | Profile feat. prompt → LLM → individual attribute [2402.12620] |
| Tabular      | Classification/regression     | Tabular rows as text → LLM (ICL) → output [2508.17391, 2508.19563] |
| Time series  | Zero-shot forecasting         | Decomposed sequence → LLM → forecast (numeric parsed) [2506.02389] |
| Event KGs    | Object/multi-event forecast   | Event quadruple/quintuple prompts → LLM [2406.10492]      |
| Performance  | NAS perf. prediction          | Hyperparam/instruction prompt → LLM → metric [2310.16712] |
| Ensemble ML  | Expert forecast weighting     | Historical expert perf. → prompt → ensemble pred. [2506.23154] |

LLMs-as-predictors are notable for two key capabilities:
- **Key-value understanding and cross-modal input reasoning** (e.g., combining numerical, categorical, and textual variables such as market indices, earnings transcripts, and analyst ratings in financial forecasting) [2408.06634].
- **Intrinsic interpretability via chain-of-thought or explanation generation** (the model can be explicitly instructed to produce explanations) [2308.15197, 2405.20962].

## 3. Strengths and Empirical Performance

LLMs-as-predictors demonstrate several empirically substantiated strengths:
- **Competitive zero- and few-shot performance for classification tasks:** For node classification, best-case accuracies on Pubmed using LLMs reach 90.75%, outperforming shallow baselines and approaching GNN performance [2307.03393].
- **Versatility across data types:** LLMs can be used without any domain-specific retraining, providing a universal baseline on small tabular datasets (classification accuracy >0.93 on several benchmarks) [2508.17391].
- **Cross-domain transferability:** LLM predictors for dynamic text-attributed graphs (using agent-based summary and reflection modules) achieve performance comparable to or exceeding fully supervised GNNs, even without dataset-specific training [2503.03258].
- **Interpretability and explanatory power:** In mobility prediction, LLMs not only rank likely next-locations but can generate interpretable reasoning about temporal and spatial regularities [2308.15197, 2405.20962].
- **Robustness to some classes of adversarial attacks:** LLMs-as-predictors, especially when incorporating neighbor summarization/fine-tuning, show lower accuracy degradation than MLPs and even some GNNs under textual/structural graph attacks [2407.12068].
- **Data augmentation and ensemble augmentation:** LLMs can serve as “soft label” makers in transfer learning—aiding ML models to adapt to covariate shift and improve out-of-distribution accuracy [2405.05445].

## 4. Limitations and Failure Modes

Despite their versatility, LLMs-as-predictors display significant weaknesses:
- **Sensitivity to prompt design and input serialization:** Tiny prompt changes (label wording, feature order, format) can cause large swings in predictive accuracy, with error deltas exceeding 80% in some tabular regression settings [2508.19563]. Variable/place name changes and row order disrupt in-context learning effects due to U-shaped attention bias toward prompt ends [2508.19563].
- **Limited utility for regression and clustering:** LLMs underperform well-tuned specialized ML models for continuous output (negative R² often observed) and fail to deliver stable results in clustering tasks [2508.17391].
- **Dependence on latent shortcut correlations for social prediction:** Without explicit shortcut features (strongly correlated demographic or affiliation indicators), LLM performance drops to near-chance in individual-level social feature prediction; they tend to default to population-level means rather than individualized inference [2402.12620].
- **Context length and scalability constraints:** The amount of historical or structural data LLMs can meaningfully attend to is capped by the model’s context window, requiring aggressive preprocessing or context consolidation [2503.03258, 2307.03393].
- **High computational and financial cost** compared to specialized models, especially for batch or streaming inference at scale.
- **Ambiguity in evaluation:** Generated outputs can appear “wrong” by ground-truth metrics while being semantically reasonable (e.g., labeling a paper “Neural Networks” instead of “Reinforcement Learning,” which might both be plausible) [2307.03393].

## 5. Methodological Innovations and Ensemble Integration

Several studies propose ways to combine LLM predictions with other models for improved reliability:
- **Linear and adaptive ensembling:** Weighted or piecewise-linear combinations of LLM and ML model predictions calibrated via cross-validation yield systematic, robust improvements in classification metrics [2405.05445].
- **Post-hoc calibration:** LLM predictions can be used to calibrate or correct probability estimates from conventional classifiers, especially in the presence of distribution shift [2405.05445].
- **Multi-model (stacked) approaches:** For rare-event prediction (e.g., VC success), LLM-powered feature engineering is used to construct and enrich feature sets, which are then fed into black-box ensemble models (XGBoost, Random Forest, Linear Regression), producing continuous predictions that are thresholded to classify rare outcomes [2509.08140].
- **Distribution-based output parsing:** Rather than treating the LLM as a point estimator, the token probability distribution is interpreted as a predictive posterior—critical for election forecasting or uncertainty quantification [2411.03486].

These innovations allow for more robust, interpretable, and often more accurate predictive pipelines, though the optimal combinations and integration logic remain context- and data-dependent.

## 6. Interpretability, Explanation, and Transparency

A distinguishing feature of LLM-based predictors—particularly when designed to produce chain-of-thought or “reasons for prediction” explanations—is their capacity for interpretability, both in mobility forecasting [2308.15197, 2405.20962] and in scenarios calling for feature attribution or decision traceability [2509.08140]. Sensitivity analysis and parameter weight inspection in multi-model frameworks further enhance transparency [2509.08140]. For distribution-based prediction, the output distribution itself provides insight into the model’s world knowledge and uncertainty, facilitating algorithmic fidelity checks and bias analysis [2411.03486].

## 7. Outlook and Research Directions

Emerging lines of research seek to improve the reliability, scalability, and applicability of LLMs-as-predictors:
- **Architectural invariance:** There is an explicit call to develop LLM architectures or training procedures that are robust to task-irrelevant variations, such as variable/row order or label serialization [2508.19563].
- **Hierarchical, multi-agent, and domain-adaptive approaches:** Multi-agent LLM frameworks with distinct roles for global and local summarization, as well as knowledge reflection, show promise for dynamic graph prediction and cross-domain adaptability without retraining [2503.03258].
- **Integration with retrieval and domain-specific modules:** LLMs may be combined with retrieval-augmented tools for overcoming context-length restraints and domain holes [2406.10492].
- **Extended explanation and self-assessment:** Advanced prompting for explicit uncertainty quantification and error analysis is being developed, with distribution-based methods enabling rich model introspection [2411.03486].
- **Application to rare-event and low-data scenarios:** LLMs combined with model ensembles and feature-engineering pipelines provide substantial gains in precision and actionable insight for rare-event prediction in VC and other high-stakes decision processes [2509.08140].

Overall, while LLMs-as-predictors supply a flexible and interpretable predictive interface across various data modalities, their practical deployment demands careful prompt engineering, robustness analysis, and often auxiliary integration with model ensembles or domain-specific processing. Robustness to task-irrelevant perturbations and calibration under out-of-distribution settings remain core challenges for future research.

Source: https://www.emergentmind.com/topics/llms-as-predictors