External Memory

Open-source foundation models have seen rapid adoption and development, enabling powerful general-purpose capabilities across diverse domains. However, fine-tuning large foundation models for domain-specific or personalized tasks remains prohibitively expensive for most users due to the significant memory overhead beyond that of inference. We introduce EMLoC, an Emulator-based Memory-efficient fine-tuning framework with LoRA Correction, which enables model fine-tuning within the same memory budget required for inference. EMLoC constructs a task-specific light-weight emulator using activation-aware singular value decomposition (SVD) on a small downstream calibration set. Fine-tuning then is performed on this lightweight emulator via LoRA. To tackle the misalignment between the original model and the compressed emulator, we propose a novel compensation algorithm to correct the fine-tuned LoRA module, which thus can be merged into the original model for inference. EMLoC supports flexible compression ratios and standard training pipelines, making it adaptable to a wide range of applications. Extensive experiments demonstrate that EMLoC outperforms other baselines across multiple datasets and modalities. Moreover, without quantization, EMLoC enables fine-tuning of a 38B model, which originally required 95GB of memory, on a single 24GB consumer GPU—bringing efficient and practical model adaptation to individual users.

Fast Fourier Transforms (FFT) are widely used to reduce memory and computational costs in deep learning. However, existing implementations, including standard FFT and real FFT (rFFT), cannot achieve true in-place computation. In particular, rFFT maps an input of size $n$ to a complex output of size $\frac{n}{2}+1$, causing dimensional mismatch and requiring additional memory allocation. We propose the first real-domain, fully in-place FFT framework (rdFFT) that preserves input-output dimensional consistency ($n \rightarrow n$). By leveraging butterfly operation symmetry and conjugate properties in the frequency domain, we design an implicit complex encoding scheme that eliminates intermediate cache usage entirely. Theoretically, our method reduces memory usage by 50\% compared to rFFTs. Moreover, it enables zero-cache parameter updates by utilizing the derivative property of the Fourier transform to compute matrix inverses efficiently without intermediate storage. Experiments on multiple natural language understanding tasks demonstrate the method’s effectiveness in maintaining model performance while significantly lowering memory overhead, offering a promising direction for frequency-domain lightweight adaptation.

Over the last decades, developing more powerful neural architectures and simultaneously designing optimization algorithms to effectively train them have been the core of research efforts to enhance the capability of machine learning models. Despite the recent progresses, particularly in developing Language Models (LMs), there are fundamental challenges and unanswered questions about how such models can continually learn/memorize, self-improved, and find ''effective solutions,''. In this paper, we present a new learning paradigm, called Nested Learning (NL), that coherently represents a model with a set of nested, multi-level, and/or parallel optimization problems, each of which with its own ''context flow''. NL reveals that existing deep learning methods learns from data through \emph{compressing} their own context flow, and explain how in-context learning emerges in large models. NL suggests a path (a new dimension to deep learning) to design more expressive learning algorithms with more ''levels'', resulting in higher-order in-context learning abilities. In addition to its neuroscientifically plausible and mathematically white-box nature, we advocate for its importance by presenting three core contributions: (1) Deep Optimizers: Based on NL, we show that well-known gradient-based optimizers (e.g., Adam, SGD with Momentum, etc.) are in fact associative memory modules that aim to compress the gradients with gradient descent. Building on this insight, we present a set of more expressive optimizers with deep memory and/or more powerful learning rules; (2) Self-Modifying Titans: Taking advantage of NL's insights on learning algorithms, we present a novel sequence model that learns how to modify itself by learning its own update algorithm; and (3) Continuum Memory System: We present a new formulation for memory system that generalizes the traditional viewpoint of ``long-term/short-term memory''. Combining our self-modifying sequence model with the continuum memory system, we present a learning module, called Hope, showing promising results in language modeling, continual learning, and long-context reasoning tasks.

Recent studies have shown LLMs possess some ability to improve their responses when given external feedback. However, it remains unclear how effectively and thoroughly these models can incorporate extrinsic feedback. In an ideal scenario, if LLMs receive near-perfect and complete feedback, we would expect them to fully integrate the feedback and reach correct solutions. In this paper, we systematically investigate LLMs’ ability to incorporate feedback by designing a controlled experimental environment. For each problem, a solver model attempts a solution, then a feedback generator with access to near-complete ground-truth answers produces targeted feedback, after which the solver tries again. We evaluate this pipeline across a diverse range of tasks, including math reasoning, knowledge reasoning, scientific reasoning, and general multi-domain evaluations with state-of-the-art language models including Claude 3.7 with extended thinking. Surprisingly, even under these near-ideal conditions, solver models consistently show resistance to feedback, a limitation that we term FEEDBACK FRICTION. To mitigate this limitation, we experiment with sampling-based strategies like progressive temperature increases and explicit rejection of previously attempted incorrect answers, which yield improvements but still fail to help models achieve target performance. We analyze FEEDBACK FRICTION and find that models’ confidence on specific questions, measured by semantic entropy, predicts feedback resistance: high-confidence predictions remain resistant to external correction. We hope that highlighting this issue in LLMs will help future research in self-improvement.

Future link prediction on temporal graphs is a fundamental task with wide applicability in real-world dynamic systems. These scenarios often involve both recurring (seen) and novel (unseen) interactions, requiring models to generalize effectively across both types of edges. However, existing methods typically rely on complex memory and aggregation modules, yet struggle to handle unseen edges. In this paper, we revisit the architecture of existing temporal graph models and identify two essential but overlooked modeling requirements for future link prediction: representing nodes with unique identifiers and performing target-aware matching between source and destination nodes. To this end, we propose Cross-Attention based Future Link Predictor on Temporal Graphs (CRAFT), a simple yet effective architecture that discards memory and aggregation modules and instead builds on two components: learnable node embeddings and cross-attention between the destination and the source's recent interactions. This design provides strong expressive power and enables target-aware modeling of the compatibility between candidate destinations and the source's interaction patterns. Extensive experiments on diverse datasets demonstrate that CRAFT consistently achieves superior performance with high efficiency, making it well-suited for large-scale real-world applications.

Organisms constantly pivot between tasks such as evading predators, foraging, traversing rugged terrain, and socializing, often within milliseconds. Remarkably, they preserve knowledge of once-learned environments sans catastrophic forgetting, a phenomenon neuroscientists hypothesize, is due to a singular neural circuitry dynamically overlayed by neuromodulatory agents such as dopamine and acetylcholine. In parallel, deep learning research addresses analogous challenges via domain generalization ($\textbf{DG}$) and continual learning ($\textbf{CL}$), yet these methods remain siloed, despite the brain’s ability to perform them seamlessly. In particular, prior work has not explored architectures involving associative memories ($\textbf{AM}$s), which are an integral part of biological systems, to jointly address these tasks. We propose Memory-Integrated Reconfigurable Adapters ($\textbf{MIRA}$), a unified framework that integrates Hopfield-style associative memory modules atop a shared backbone. These memory modules store adapter-weight updates as values and retrieve them via learned keys. Associative memory keys are learned post-hoc to index and retrieve an affine combination of stored adapter updates for any given task or domain on a per-sample basis. By varying only the task-specific objectives, we demonstrate that $\textbf{MIRA}$ seamlessly accommodates domain shifts and sequential task exposures under one roof. Empirical evaluations on standard benchmarks confirm that our $\textbf{AM}$-augmented architecture significantly enhances adaptability and retention: in $\textbf{DG}$, $\textbf{MIRA}$ achieves SoTA out-of-distribution accuracy, and in incremental learning settings, it outperforms architectures explicitly designed to handle catastrophic forgetting using generic $\textbf{CL}$ algorithms. Extensive ablation studies validate the necessity of both associative memory storage and post-hoc key learning for robust interpolated retrieval of adapters. By unifying adapter-based modulation with biologically inspired associative memory, $\textbf{MIRA}$ delivers rapid task switching and enduring knowledge retention in a single extensible architecture, charting a path toward more versatile and memory-augmented AI systems.

We propose a novel energy function for Dense Associative Memory (DenseAM) networks, the log-sum-ReLU (LSR), inspired by optimal kernel density estimation. Unlike the common log-sum-exponential (LSE) function, LSR is based on the Epanechnikov kernel and enables exact memory retrieval with exponential capacity without requiring exponential separation functions. Uniquely, it introduces abundant additional emergent local minima while preserving perfect pattern recovery --- a characteristic previously unseen in DenseAM literature. Empirical results show that LSR energy has significantly more local minima (memories) that have comparable log-likelihood to LSE-based models. Analysis of LSR's emergent memories on image datasets reveals a degree of creativity and novelty, hinting at this method's potential for both large-scale memory storage and generative tasks.

Establishing object-level correspondence between egocentric and exocentric views is essential for intelligent assistants to deliver precise and intuitive visual guidance. However, this task faces numerous challenges, including extreme viewpoint variations, occlusions, and the presence of small objects. Existing approaches usually borrow solutions from video object segmentation models, but still suffer from the aforementioned challenges. Recently, the Segment Anything Model 2 (SAM 2) has shown strong generalization capabilities and excellent performance in video object segmentation. Yet, when simply applied to the ego-exo correspondence (EEC) task, SAM 2 encounters severe difficulties due to ineffective ego-exo feature fusion and limited long-term memory capacity, especially for long videos. Addressing these problems, we propose a novel EEC framework based on SAM 2 with long-term memories by presenting a dual-memory architecture and an adaptive feature routing module inspired by Mixture-of-Experts (MoE). Compared to SAM 2, our approach features **(i)** a Memory-View MoE module which consists of a dual-branch routing mechanism to adaptively assign contribution weights to each expert feature along both channel and spatial dimensions, and **(ii)** a dual-memory bank system with a simple yet effective compression strategy to retain critical long-term information while eliminating redundancy. In the extensive experiments on the challenging EgoExo4D benchmark, our method, dubbed ***LM-EEC***, achieves new state-of-the-art results and significantly outperforms existing methods and the SAM 2 baseline, showcasing its strong generalization across diverse scenarios. Our code and model are available at https://github.com/juneyeeHu/LM-EEC.

Memory Mosaics, networks of associative memories, have demonstrated appealing compositional and in-context learning capabilities on medium-scale networks (GPT-2 scale) and synthetic small datasets. This work shows that these favorable properties remain when we scale memory mosaics to large language model sizes (llama-8B scale) and real-world datasets. To this end, we scale memory mosaics to 10B size, we train them on one trillion tokens, we introduce a couple architectural modifications (*memory mosaics v2*), we assess their capabilities across three evaluation dimensions: training-knowledge storage, new-knowledge storage, and in-context learning. Throughout the evaluation, memory mosaics v2 match transformers on the learning of training knowledge (first dimension) and significantly outperforms transformers on carrying out new tasks at inference time (second and third dimensions). These improvements cannot be easily replicated by simply increasing the training data for transformers. A memory mosaics v2 trained on one trillion tokens still perform better on these tasks than a transformer trained on eight trillion tokens.

While large language model (LLM) agents can effectively use external tools for complex real-world tasks, they require memory systems to leverage historical experiences. Current memory systems enable basic storage and retrieval but lack sophisticated memory organization, despite recent attempts to incorporate graph databases. Moreover, these systems' fixed operations and structures limit their adaptability across diverse tasks. To address this limitation, this paper proposes a novel agentic memory system for LLM agents that can dynamically organize memories in an agentic way. Following the basic principles of the Zettelkasten method, we designed our memory system to create interconnected knowledge networks through dynamic indexing and linking. When a new memory is added, we generate a comprehensive note containing multiple structured attributes, including contextual descriptions, keywords, and tags. The system then analyzes historical memories to identify relevant connections, establishing links where meaningful similarities exist. Additionally, this process enables memory evolution -- as new memories are integrated, they can trigger updates to the contextual representations and attributes of existing historical memories, allowing the memory network to continuously refine its understanding. Our approach combines the structured organization principles of Zettelkasten with the flexibility of agent-driven decision making, allowing for more adaptive and context-aware memory management. Empirical experiments on six foundation models show superior improvement against existing SOTA baselines. The code is available at \url{https://anonymous.4open.science/r/AgenticMemory-76B4}.

Current Large Language Models (LLMs) are confronted with overwhelming information volume when comprehending long-form documents. This challenge raises the imperative of a cohesive memory module, which can elevate vanilla LLMs into autonomous reading agents. Despite the emergence of some heuristic approaches, a systematic design principle remains absent. To fill this void, we draw inspiration from Jean Piaget's Constructivist Theory, illuminating three traits of the agentic memory---structured schemata, flexible assimilation, and dynamic accommodation. This blueprint forges a clear path toward a more robust and efficient memory system for LLM-based reading comprehension. To this end, we develop CAM, a prototype implementation of Constructivist Agentic Memory that simultaneously embodies the structurality, flexibility, and dynamicity. At its core, CAM is endowed with an incremental overlapping clustering algorithm for structured memory development, supporting both coherent hierarchical summarization and online batch integration. During inference, CAM adaptively explores the memory structure to activate query-relevant information for contextual response, akin to the human associative process. Compared to existing approaches, our design demonstrates dual advantages in both performance and efficiency across diverse long-text reading comprehension tasks, including question answering, query-based summarization, and claim verification.

Recent works show that contrastive learning based on memory banks is an effective framework for unsupervised person Re-IDentification (ReID). In existing methods, memory banks are typically initialized with cluster centroids and rewritten with positive samples via the momentum mechanism along with the model training. However, this mechanism solely focuses on the intra-class compactness by pulling memory banks close to positive samples, neglecting the inter-class separability among different memory banks. Rewriting memory banks with partial constraint limits their discrimination capacities, and hence hinders learning discriminative features based on those memory banks. In this paper, we claim that memory banks should be rewritten with both intra-class and inter-class constraints, and therefore propose a unified memory bank rewriting mechanism, Bidirectionally Memory bank reWriting (BMW), to chase enhanced discrimination capacity. Specifically, BMW formulates the memory bank rewriting as the gradient descent update with two objectives, i.e., reducing intra-class diversity and enhancing inter-class separability. To effectively enhance the separability of memory banks with limited number of rewriting steps, we further design a novel objective formulation for the inter-class constraint, which is more effective for one step update. BMW enhances both representation and discrimination capacities of memory banks, thus leads to an effective ReID feature optimization. BMW is simple yet effective and can serve as a new paradigm for person ReID methods based on memory banks. Extensive experiments on standard benchmarks demonstrate the effectiveness of our BMW method in unsupervised ReID model training. Specially, BMW even outperforms previous methods that use stronger backbones. Code is available at https://github.com/liu-xb/BMW.