Knowledge catalogue
This IBM Research article discusses optimizing AI inference performance and costs by leveraging mixed hardware configurations, combining different types of processors and accelerators rather than rely
arXiv:2606.22296v1 Announce Type: new Abstract: Edge Internet of Things (IoT) agents are often constrained by memory capacity, privacy requirements, communication latency, and recurring inference cost
arXiv:2511.20889v2 Announce Type: replace Abstract: Test-time alignment (TTA) aims to adapt models to specific rewards during inference. However, existing methods tend to either under-optimise or over
arXiv:2606.21185v1 Announce Type: cross Abstract: There is a precise sense in which drawing causal inferences from observational data is hard, even when identifiability is assumed. In particular, Robi
b9763 is a release of llama.cpp, an open-source LLM inference project in C/C++ . The release includes built binaries across multiple platforms including macOS, Linux, Windows, Android, and openEuler,
Release b9752 of llama.cpp focused on refactoring batch construction in the server component (PR #24843) , implementing improvements to how inference batches are handled. The release includes builds f
The Top AI Papers of the Week (June 14 - June 21): - PreAct - SpatialClaw - Back on Track - OpenClaw-Skill - From Trainee to Trainer - Compositional Skill Routing - Can LLM Agents Infer World Models?
This is what the AI engineering ecosystem looks like in one room. 60+ Gold Sponsors are joining us at AI Engineer World's Fair, spanning infra, inference, agents, data, eval, and observability. The wh
arXiv:2603.21639v2 Announce Type: replace-cross Abstract: Accurately estimating human mobility in peripheral regional economies presents a fundamental measurement challenge: physical ground-truth sens
arXiv:2606.11634v1 Announce Type: new Abstract: The rapid progress of reasoning and agentic large language models (LLMs) has increased the demand for long-context inference, but self-attention (SA) sc
arXiv:2606.12279v1 Announce Type: cross Abstract: Recent work in ML applies genetic algorithms at inference time to iteratively improve solutions to optimization problems. The basic mutation and recom
arXiv:2606.11645v1 Announce Type: new Abstract: Micro-gesture analysis attracts increasing attention for inferring spontaneous emotion from subtle body movements. Micro-gesture online recognition, whi
arXiv:2606.12412v1 Announce Type: cross Abstract: Vision-language models (VLMs) project images into hundreds to thousands of visual tokens, making decoder inference expensive in both attention computa
arXiv:2606.11543v1 Announce Type: new Abstract: Agent Skills augment large language model (LLM) agents with procedural knowledge at inference time, but current benchmarks rarely distinguish what a Ski
arXiv:2606.12316v1 Announce Type: new Abstract: ARC tests in-context rule induction: given a few input-output demonstrations, a model must infer the hidden rule and apply it to a new query. While many
arXiv:2606.11770v1 Announce Type: new Abstract: Spatial reasoning remains a challenge for Multimodal Large Language Models (MLLMs), as it requires reliable multi-hop inference over both intermediate s
arXiv:2606.11269v1 Announce Type: new Abstract: Personality assessment aims to infer stable personality traits from dynamic behaviors across language, voice, and facial cues. Since different personali
arXiv:2606.11599v1 Announce Type: new Abstract: Activation steering offers a lightweight approach to control language models' behavior at inference time, but whether it succeeds or fails heavily depen
arXiv:2409.04111v2 Announce Type: replace Abstract: Vertical federated learning is a natural and elegant approach to integrate multi-view data vertically partitioned across devices (clients) while pre
arXiv:2606.09864v1 Announce Type: cross Abstract: Key-value (KV) cache quantization is widely used to reduce Large Language Model (LLM) inference memory, yet existing evaluations solely focus on measu
arXiv:2606.10440v1 Announce Type: cross Abstract: Distributed machine learning (ML) is a key paradigm for today's large-scale artificial intelligence applications. As model inference arises as an impo
arXiv:2606.10829v1 Announce Type: cross Abstract: Masked diffusion language models can reduce inference steps by revealing multiple tokens per denoising iteration, but this parallelism is fragile: pos
arXiv:2606.09919v1 Announce Type: cross Abstract: Perceptual uncertainty is a central challenge for heterogeneous robot teams operating in unstructured outdoor environments, where no single viewpoint
arXiv:2510.04195v2 Announce Type: replace Abstract: Given a map description through global traversal navigation instructions, an LLM can often infer the implicit spatial layout and answer user queries
arXiv:2606.10040v1 Announce Type: new Abstract: World-Action Models (WAMs) have emerged as a promising paradigm for embodied control by coupling future visual prediction with action generation. Howeve
arXiv:2606.09946v1 Announce Type: cross Abstract: Edge-AI systems increasingly require real-time CNN inference under strict energy, performance, security, and privacy constraints. Approximate computin
arXiv:2606.09877v1 Announce Type: cross Abstract: LLM wiki systems compile knowledge into pre-filled KV caches for efficient inference, but assume a static corpus -- an assumption that fails whenever
arXiv:2606.10371v1 Announce Type: cross Abstract: Diffusion-based action generation has become a foundational component of embodied AI, but its reliance on visual conditioning leaves deployed visuomot
arXiv:2606.09856v1 Announce Type: cross Abstract: Post-training Large Language Models (LLMs) for reasoning typically focuses on deductive tasks such as mathematics and coding where correctness is veri
arXiv:2606.10423v1 Announce Type: new Abstract: Autonomous web navigation remains challenging for LLM agents, and the strongest generalist systems rely on proprietary reasoning models whose inference
arXiv:2606.08336v1 Announce Type: new Abstract: While multimodal integration significantly improves computer vision models, deploying them incurs prohibitive inference costs and requires scarce, perfe
As enterprise AI budgets hit their limits earlier each year, the pressure to fund new agentic and inference workloads without expanding total spend is forcing a fundamental rethink of infrastructure a
arXiv:2606.07646v1 Announce Type: cross Abstract: Test-time adaptation (TTA) aims to align a model to shifting test domains using only unlabeled streaming data. Most existing methods implicitly infer
arXiv:2606.07915v1 Announce Type: new Abstract: Neural symbolic regression models improve inference efficiency by shifting structural search to pretraining, but their one-pass autoregressive decoding
arXiv:2606.08658v1 Announce Type: new Abstract: LLMs have revolutionized knowledge representation and retrieval, but lack the explicit modeling that knowledge ontologies possess. This paper surveys th
arXiv:2606.07586v1 Announce Type: cross Abstract: Spatial neural processing units (NPUs) provide an energy-efficient platform for edge LLM inference, but efficiently deploying an LLM end-to-end on suc
arXiv:2504.05349v4 Announce Type: replace-cross Abstract: Network pruning is used to reduce inference latency and power consumption in large neural networks. However, most methods focus on empirical r
arXiv:2506.20573v4 Announce Type: replace-cross Abstract: Public datasets, crucial for modern machine learning and statistical inference, often contain low-quality or contaminated samples that can har
arXiv:2606.07577v1 Announce Type: new Abstract: Audio-visual large language models (LLMs) hold strong promise for long-form video understanding, yet their long-video inference is fundamentally limited
arXiv:2606.09189v1 Announce Type: cross Abstract: EEG foundation-model releases are usually audited one endpoint at a time: raw-reconstruction, membership inference, identity linkage, or DP-SGD on the
arXiv:2606.08804v1 Announce Type: new Abstract: Linear attention reformulates sequence modeling as recurrent state evolution, enabling efficient linear-time inference. Under the key-value associative
arXiv:2606.08779v1 Announce Type: new Abstract: Reinforcement Learning (RL) has emerged as a pivotal post-training paradigm, yet it frequently suffers from unpredictable sub-optimum performance or eve
arXiv:2606.09367v1 Announce Type: new Abstract: In real-world deployment under strict real-time constraints, weather and imaging variations induce significant distribution shifts, severely degrading d
arXiv:2606.08402v1 Announce Type: cross Abstract: Generating complete 3D scenes from a single image requires inferring globally consistent geometry, object relationships, and environmental context fro
arXiv:2606.07684v1 Announce Type: cross Abstract: Disaggregated serving alleviates memory bottlenecks in Large Language Model (LLM) inference but creates a severe communication bottleneck: transmittin
arXiv:2507.09751v3 Announce Type: replace Abstract: Large language models (LLMs) have demonstrated impressive capabilities in natural language understanding and generation, but exhibit problems with l
arXiv:2606.09135v1 Announce Type: cross Abstract: We demonstrate that widely deployed Large Language Model (LLM) inference stacks harbor a steganographic channel that requires no modification to model
arXiv:2510.16028v4 Announce Type: replace-cross Abstract: Neural networks increasingly run on hardware outside the user's control (cloud GPUs, inference marketplaces). Yet ML-as-a-Service reveals litt
arXiv:2606.07895v1 Announce Type: new Abstract: Discrete Vision-Language-Action (VLA) models typically formulate action generation as next-token prediction over discretized action spaces, conditioning
arXiv:2606.08122v1 Announce Type: new Abstract: Predicting a user's next Point-of-Interest (POI) based on their historical check-in records is a fundamental task in location-based services. While rece
arXiv:2606.08408v1 Announce Type: cross Abstract: We extend activation steering to diffusion language models (DLMs) and study a novel problem that arose due to the inference mechanism of DLMs: Modifyi
arXiv:2606.09474v1 Announce Type: new Abstract: Generalized Few-Shot Semantic Segmentation (GFSS) has traditionally been approached as a representation-learning problem, requiring task-specific adapta
arXiv:2601.22736v2 Announce Type: replace-cross Abstract: Causal inference from observational data can provide strong evidence for finding the best action in a decision-making scenario without having
arXiv:2606.08202v1 Announce Type: cross Abstract: Most statistical and machine learning methods for directed interactions focus on pairwise effects among variables. Even existing cyclic models represe
arXiv:2606.07410v1 Announce Type: cross Abstract: The emergence of 'Aha moments' in large language models, particularly DeepSeek-R1-0120, has raised the question of whether these systems genuinely rea
arXiv:2606.06946v1 Announce Type: cross Abstract: We present LoRA-MINT, a new methodology for Membership Inference Test (MINT) applied to recent Large Language Models (LLMs) fine-tuned for specific Na
llama.cpp is an LLM inference tool in C/C++ , and b9555 represents a specific build or version release from the ggml-org/llama.cpp GitHub repository. The project does not follow traditional release pr
B9561 is an intermediate build release of llama.cpp, the C/C++ implementation of large language model inference. Llama.cpp releases use build identifiers (b-numbers) to track development versions betw
b9563 is a release build of llama.cpp, an open-source software library for large language model inference that is co-developed alongside the GGML tensor library. This intermediate build includes vario
arXiv:2606.06712v1 Announce Type: cross Abstract: We study the transformation of autoregressive models (ARLMs) into diffusion language models (DLMs). Rather than pretraining from scratch, prior work r