CL-bench: A Benchmark for Context Learning

Current language models (LMs) excel at reasoning over prompts using pre-trained knowledge. However, real-world tasks are far more complex and context-dependent: models must learn from task-specific context and leverage new knowledge beyond what is learned during pre-training to reason and resolve tasks. We term this capability context learning, a crucial ability that humans naturally possess but has been largely overlooked. To this end, we introduce CL-bench, a real-world benchmark consisting of 500 complex contexts, 1,899 tasks, and 31,607 verification rubrics, all crafted by experienced domain experts. Each task is designed such that the new content required to resolve it is contained within the corresponding context. Resolving tasks in CL-bench requires models to learn from the context, ranging from new domain-specific knowledge, rule systems, and complex procedures to laws derived from empirical data, all of which are absent from pre-training. This goes far beyond long-context tasks that primarily test retrieval or reading comprehension, and in-context learning tasks, where models learn simple task patterns via instructions and demonstrations. Our evaluations of ten frontier LMs find that models solve only 17.2% of tasks on average. Even the best-performing model, GPT-5.1, solves only 23.7%, revealing that LMs have yet to achieve effective context learning, which poses a critical bottleneck for tackling real-world, complex context-dependent tasks. CL-bench represents a step towards building LMs with this fundamental capability, making them more intelligent and advancing their deployment in real-world scenarios.

Efficient Estimation of Kernel Surrogate Models for Task Attribution

Modern AI agents such as large language models are trained on diverse tasks -- translation, code generation, mathematical reasoning, and text prediction -- simultaneously. A key question is to quantify how each individual training task influences performance on a target task, a problem we refer to as task attribution. The direct approach, leave-one-out retraining, measures the effect of removing each task, but is computationally infeasible at scale. An alternative approach that builds surrogate models to predict a target task's performance for any subset of training tasks has emerged in recent literature. Prior work focuses on linear surrogate models, which capture first-order relationships, but miss nonlinear interactions such as synergy, antagonism, or XOR-type effects. In this paper, we first consider a unified task weighting framework for analyzing task attribution methods, and show a new connection between linear surrogate models and influence functions through a second-order analysis. Then, we introduce kernel surrogate models, which more effectively represent second-order task interactions. To efficiently learn the kernel surrogate, we develop a gradient-based estimation procedure that leverages a first-order approximation of pretrained models; empirically, this yields accurate estimates with less than $2\%$ relative error without repeated retraining. Experiments across multiple domains -- including math reasoning in transformers, in-context learning, and multi-objective reinforcement learning -- demonstrate the effectiveness of kernel surrogate models. They achieve a $25\%$ higher correlation with the leave-one-out ground truth than linear surrogates and influence-function baselines. When used for downstream task selection, kernel surrogate models yield a $40\%$ improvement in demonstration selection for in-context learning and multi-objective reinforcement learning benchmarks.

VIRAL: Visual In-Context Reasoning via Analogy in Diffusion Transformers

Replicating In-Context Learning (ICL) in computer vision remains challenging due to task heterogeneity. We propose \textbf{VIRAL}, a framework that elicits visual reasoning from a pre-trained image editing model by formulating ICL as conditional generation via visual analogy ($x_s : x_t :: x_q : y_q$). We adapt a frozen Diffusion Transformer (DiT) using role-aware multi-image conditioning and introduce a Mixture-of-Experts LoRA to mitigate gradient interference across diverse tasks. Additionally, to bridge the gaps in current visual context datasets, we curate a large-scale dataset spanning perception, restoration, and editing. Experiments demonstrate that VIRAL outperforms existing methods, validating that a unified V-ICL paradigm can handle the majority of visual tasks, including open-domain editing. Our code is available at https://anonymous.4open.science/r/VIRAL-744A

A Minimal Task Reveals Emergent Path Integration and Object-Location Binding in a Predictive Sequence Model

Adaptive cognition requires structured internal models representing objects and their relations. Predictive neural networks are often proposed to form such "world models", yet their underlying mechanisms remain unclear. One hypothesis is that action-conditioned sequential prediction suffices for learning such world models. In this work, we investigate this possibility in a minimal in-silico setting. Sequentially sampling tokens from 2D continuous token scenes, a recurrent neural network is trained to predict the upcoming token from current input and a saccade-like displacement. On novel scenes, prediction accuracy improves across the sequence, indicating in-context learning. Decoding analyses reveal path integration and dynamic binding of token identity to position. Interventional analyses show that new bindings can be learned late in sequence and that out-of-distribution bindings can be learned. Together, these results demonstrate how structured representations that rely on flexible binding emerge to support prediction, offering a mechanistic account of sequential world modeling relevant to cognitive science.

They Said Memes Were Harmless-We Found the Ones That Hurt: Decoding Jokes, Symbols, and Cultural References

Meme-based social abuse detection is challenging because harmful intent often relies on implicit cultural symbolism and subtle cross-modal incongruence. Prior approaches, from fusion-based methods to in-context learning with Large Vision-Language Models (LVLMs), have made progress but remain limited by three factors: i) cultural blindness (missing symbolic context), ii) boundary ambiguity (satire vs. abuse confusion), and iii) lack of interpretability (opaque model reasoning). We introduce CROSS-ALIGN+, a three-stage framework that systematically addresses these limitations: (1) Stage I mitigates cultural blindness by enriching multimodal representations with structured knowledge from ConceptNet, Wikidata, and Hatebase; (2) Stage II reduces boundary ambiguity through parameter-efficient LoRA adapters that sharpen decision boundaries; and (3) Stage III enhances interpretability by generating cascaded explanations. Extensive experiments on five benchmarks and eight LVLMs demonstrate that CROSS-ALIGN+ consistently outperforms state-of-the-art methods, achieving up to 17% relative F1 improvement while providing interpretable justifications for each decision.

From Zero to Hero: Advancing Zero-Shot Foundation Models for Tabular Outlier Detection

Outlier detection (OD) is widely used in practice; but its effective deployment on new tasks is hindered by lack of labeled outliers, which makes algorithm and hyperparameter selection notoriously hard. Foundation models (FMs) have transformed ML, and OD is no exception: Shen et. al. (2025) introduced FoMo-0D, the first FM for OD, achieving remarkable performance against numerous baselines. This work introduces OUTFORMER, which advances FoMo-0D with (1) a mixture of synthetic priors and (2) self-evolving curriculum training. OUTFORMER is pretrained solely on synthetic labeled datasets and infers test labels of a new task by using its training data as in-context input. Inference is fast and zero-shot, requiring merely forward pass and no labeled outliers. Thanks to in-context learning, it requires zero additional work-no OD model training or bespoke model selection-enabling truly plug-and-play deployment. OUTFORMER achieves state-of-the-art performance on the prominent AdBench, as well as two new large-scale OD benchmarks that we introduce, comprising over 1,500 datasets, while maintaining speedy inference.

Mitigating Conversational Inertia in Multi-Turn Agents

Large language models excel as few-shot learners when provided with appropriate demonstrations, yet this strength becomes problematic in multiturn agent scenarios, where LLMs erroneously mimic their own previous responses as few-shot examples. Through attention analysis, we identify conversational inertia, a phenomenon where models exhibit strong diagonal attention to previous responses, which is associated with imitation bias that constrains exploration. This reveals a tension when transforming few-shot LLMs into agents: longer context enriches environmental feedback for exploitation, yet also amplifies conversational inertia that undermines exploration. Our key insight is that for identical states, actions generated with longer contexts exhibit stronger inertia than those with shorter contexts, enabling construction of preference pairs without environment rewards. Based on this, we propose Context Preference Learning to calibrate model preferences to favor low-inertia responses over highinertia ones. We further provide context management strategies at inference time to balance exploration and exploitation. Experimental results across eight agentic environments and one deep research scenario validate that our framework reduces conversational inertia and achieves performance improvements.

Counting Hypothesis: Potential Mechanism of In-Context Learning

In-Context Learning (ICL) indicates that large language models (LLMs) pretrained on a massive amount of data can learn specific tasks from input prompts' examples. ICL is notable for two reasons. First, it does not need modification of LLMs' internal structure. Second, it enables LLMs to perform a wide range of tasks/functions with a few examples demonstrating a desirable task. ICL opens up new ways to utilize LLMs in more domains, but its underlying mechanisms still remain poorly understood, making error correction and diagnosis extremely challenging. Thus, it is imperative that we better understand the limitations of ICL and how exactly LLMs support ICL. Inspired by ICL properties and LLMs' functional modules, we propose 1the counting hypothesis' of ICL, which suggests that LLMs' encoding strategy may underlie ICL, and provide supporting evidence.

Context Learning for Multi-Agent Discussion

Multi-Agent Discussion (MAD) has garnered increasing attention very recently, where multiple LLM instances collaboratively solve problems via structured discussion. However, we find that current MAD methods easily suffer from discussion inconsistency, LLMs fail to reach a coherent solution, due to the misalignment between their individual contexts.In this paper, we introduce a multi-LLM context learning method (M2CL) that learns a context generator for each agent, capable of dynamically generating context instructions per discussion round via automatic information organization and refinement. Specifically, inspired by our theoretical insights on the context instruction, M2CL train the generators to control context coherence and output discrepancies via a carefully crafted self-adaptive mechanism.It enables LLMs to avoid premature convergence on majority noise and progressively reach the correct consensus. We evaluate M2CL on challenging tasks, including academic reasoning, embodied tasks, and mobile control. The results show that the performance of M2CL significantly surpasses existing methods by 20%--50%, while enjoying favorable transferability and computational efficiency.

Language Steering for Multilingual In-Context Learning

While multilingual large language models have gained widespread adoption, their performance on non-English languages remains substantially inferior to English. This disparity is particularly evident in in-context learning scenarios, where providing demonstrations in English but testing on non-English inputs leads to significant performance degradation. In this paper, we hypothesize that LLMs develop a universal semantic space for understanding languages, where different languages are encoded as distinct directions within this space. Based on this hypothesis, we propose language vectors -- a training-free language steering approach that leverages activation differences between source and target languages to guide model behavior. We steer the model generations by adding the vector to the intermediate model activations during inference. This is done to make the model's internal representations shift towards the target language space without any parameter updates. We evaluate our method across three datasets and test on a total of 19 languages on three different models. Our results show consistent improvements on multilingual in-context learning over baselines across all tasks and languages tested. Beyond performance gains, hierarchical clustering of steering vectors reveals meaningful linguistic structure aligned with language families. These vectors also successfully transfer across tasks, demonstrating that these representations are task-agnostic.