Aidan Bradshaw, Marco Giordano, David Rode +8cs.CV
The 3D center of mass (CoM) is a primary quantity in the biomechanical analysis of sport, rehabilitation, and clinical movement, yet existing 3D pose tracking, mesh recovery, and multi-view triangulation methods either optimize 3D keypoint accuracy without anatomical constraints or carry compute and capture infrastructure too heavy to deploy where CoM tracking is most useful. As a result, the metric CoM remains difficult for coaches and movement analysts to measure from a single camera where athletes train and compete. In this work, we introduce MuyBridge, an on-device system that estimates the athlete's segmental center of mass trajectory from a single phone camera video stream. MuyBridge couples a compact 2D pose network and a distilled single-step monocular depth network through an analytic metric fusion that uses anatomical and physical priors to anchor the metric CoM, requiring no 3D or task-specific supervision. Evaluated on the athletic movements of AthletePose3D (running, track and field, and figure skating), MuyBridge achieves 33-41 mm vertical CoM error and 2.3-6.6% absolute-relative range error (AbsRel) under a one-time calibration, and produces CoM estimates at the 63 FPS pose-estimation rate using asynchronous 2.86 Hz depth updates on iPhone 15. Code is available at: https://github.com/Abradshaw1/Muybridge
The use of multimodal LLMs (MLLMs) for egocentric video understanding with wearable devices is constrained by the token budget. Memory and compute cost scale with the number of visual tokens, and high-resolution video quickly becomes expensive to transmit and process at scale. Prior work (GazeLLM) addresses this by cropping the video around the camera wearer's gaze. This reduces the number of visual tokens by about tenfold while maintaining or improving the quality of full-resolution descriptions. However, this compression strategy depends on dedicated eye-tracking hardware, which is unavailable on consumer smart glasses. Building a software-only substitute poses a joint constraint: the predictor must be accurate enough to preserve downstream description quality, yet light enough to run on-device, within the power and compute budget of a smartphone. We address this with EgoGazeLite, a lightweight dual-process gaze predictor for egocentric video. Across two MLLMs, three automated metrics, and two LLM judges, predicted-gaze crops show no significant difference from ground-truth-gaze crops. Equivalence is confirmed in all ten cases. EgoGazeLite achieves this at 15.7M parameters, 6.71 GFLOPs, and runs the full gaze-and-crop pipeline end-to-end in real time (21.6 ms/frame) on consumer accelerator hardware. Together, these results remove the need for eye-tracking hardware for token-efficient, gaze-conditioned egocentric video understanding with MLLMs.
Large language models can infer sensitive personal attributes, such as age, location, and occupation, from ordinary text, turning everyday writing into a privacy risk. Adversarial anonymization defends against this by rewriting a text with a capable language model that also plays the attacker, but it needs a powerful model at inference time and thus sends private text to a third party, the very exposure anonymization should prevent. Recent work distills this behavior into a small on-device model using supervised fine-tuning and direct preference optimization (DPO), but DPO only imitates the teacher's offline choices and never directly optimizes the privacy--utility objective we care about. We introduce \textbf{GRASP} (\textbf{G}roup-\textbf{R}elative \textbf{A}nonymization via \textbf{S}elf-refinement \textbf{P}olicy-optimization), which reinforces the local anonymizer online with Group Relative Policy Optimization. A single small model acts as anonymizer, adversary, and utility judge, trained against a self-generated reward that hides attributes while preserving meaning, with a design that guards against reward hacking. Trained on Llama-3.1-8B, \ours{} improves the privacy--utility trade-off over the DPO-distilled baseline, consistently across three independent LLM judges. Against adversarial anonymization driven by frontier models such as Gemini~2.5~Flash and Claude, it achieves a comparable or better overall trade-off while removing substantially more private information, and it runs entirely on-device at roughly $1\%$ of the GPT-4o teacher's cost.
Jiaqi Gan, Haoyuan Tang, Jamey Z. Liang +6cs.CR cs.AI
Language models small enough to run on a handset, quantized to a few bits, are increasingly capable of acting on their user's behalf -- which makes on-device task automation newly plausible. One such task is answering the phone. A phone secretary takes an unknown inbound call on its owner's behalf, and unlike the agents most benchmarks evaluate, it has no cooperative caller-assigned task to complete: the caller holds the goal and may be an adversary, while the secretary must begin deciding how to respond without an oracle. What matters is not task success but whether the owner would endorse how their proxy handled the call. We evaluate only the text-domain conversational decision layer; speech recognition, audio interaction, end-to-end latency, and handset execution are outside scope. We present CallScreenBench, which reports five automated call-and-note measure groups motivated by owner endorsement. Each is paired, where available, with a counter-metric and an uncertainty estimate; no benchmark-wide Q1-Q5 composite or leaderboard score is defined. Three guardedness diagnostics identify candidate cases for a toolless proxy that holds no credentials and calls no tools. Across three model families represented by paired 4-bit checkpoints (0.6-4B), the primary scoring snapshot gives the larger checkpoint higher point estimates on several service, recall, and plausibility measures, while triage discrimination follows a different ordering. Bare scam-side TPR rewards universal suspicion, and pairwise separation changes when legitimate-side false positives are included and across judge snapshots. Scripted degenerate agents expose further floors, including a hangup-and-echo policy with entity recall 1.000. We report quality measures and guardedness channels separately so that a single pass/fail score does not hide their trade-offs.
On-device in-context learning (ICL) relies on pre-inference retrieval to select demonstrations for useful context before downstream model inference. This retrieval must exploit task-specific information while operating over local memories under limited computation, memory, and data-exposure budgets. We propose Conditional Retrieval Alignment (CoRA), a gradient-free framework that converts a frozen encoder into a task-conditioned retriever using paired candidate inputs and outputs. CoRA selects complementary encoder layers, constructs an output-derived conditioning space from candidate memory, and aligns candidate input representations to this space through closed-form ridge regression. Low-rank factorization then produces a compact retrieval basis where candidate outputs are used only during offline index construction, whereas query-time retrieval requires only the query input and precomputed index. We show that CoRA's rank-constrained basis is the optimal low-rank compression of the output-conditioned fitted representation, and derive an exact two-pass streaming construction that avoids materializing the full fitted matrix. We further extend the framework to multimodal exemplar retrieval by incorporating visual representations into the conditioning and retrieval spaces. Experiments across ten textual datasets and four multimodal benchmarks with Llama-3.2-1B, MobileLLM-Pro, OpenFlamingo-3B, and Qwen3.5-2B, as well as end-to-end Raspberry Pi~5 deployment demonstrate that CoRA supports effective task-conditioned retrieval without retriever fine-tuning, backpropagation, or target-model calls.
Large language models (LLMs) have demonstrated remarkable capabilities in language understanding, reasoning, and world knowledge. As embodied agents become increasingly capable, there is a growing demand for compact models that can serve as an on-device brain, preserving the broad general intelligence of LLMs while enabling effective high-level interaction with embodied environments. Existing approaches, however, often prioritize either general-purpose intelligence or specialized embodied capabilities, making it challenging to satisfy both requirements within a single model. We present \textbf{Athena-Brain-8B}, an 8B LLM designed to serve as an on-device brain for embodied intelligence for embodied intelligence. Through a multi-stage post-training pipeline consisting of General Supervised Fine-Tuning, General Reinforcement Learning, Embodied Expert training, and Model Merge, Athena-Brain-8B maintains strong general capabilities while acquiring strong high-level embodied interaction capabilities and generating concise responses for efficient embodied interaction. Experimental results demonstrate the effectiveness of Athena across both general and embodied evaluations. Compared with the corresponding Qwen3-8B thinking model, Athena-Brain-8B achieves comparable performance on general language and reasoning benchmarks while generating substantially shorter responses. On in-domain embodied benchmarks, Athena-Brain-8B consistently outperforms models of similar scale and surpasses several substantially larger frontier models evaluated zero-shot, demonstrating that compact language models can effectively integrate strong general intelligence with embodied capabilities.
Himel Dev, Tanmoy Sen, Madhusudan Basak +1cs.LG cs.AI
Generating personalized trip itineraries is a complex planning task and involves a tension between hard combinatorial feasibility and soft latent desirability. Classical optimization enforces constraints but fails to capture subjective traveler preferences. While learning-based approaches model preferences, they cannot guarantee feasibility. Mobile deployment imposes additional resource constraints on both. To address this, we propose Plan, Learn, Adapt (PLA), a three-stage framework for personalized on-device itinerary generation. The Plan stage builds a heterogeneous ensemble of lightweight planners that produces structurally diverse feasible candidates. From pairwise itinerary comparisons, Learn fits a compact Bradley-Terry reward model that captures emergent schedule properties such as pacing, geographic coherence, and day balance, which per-POI signals miss. Finally, Adapt applies feasibility-preserving local refinement within a device-aware compute budget; every intermediate state is feasible by construction. On 2,519 pairwise human comparisons across more than 100 U.S. cities, the reward-guided ensemble achieves a 67.8% win rate, 11.2 percentage points above the best single planner, with 100% feasibility. Three frontier LLMs, GPT-5, Claude Opus 4.5, and Gemini 3 Pro, achieve 0% feasibility under the same constraints. The reward model generalizes across held-out cities, with a 67.6% mean leave-one-city-out accuracy. In production deployment within FlyEnJoy, PLA increased itinerary completion rates by 91%, with 109.9 ms average on-device latency.
Large Language Model (LLM) agents have moved beyond generating responses to executing multi-step tasks by calling tools, observing the results, and iteratively deciding the next action. Most agent systems run on desktops or servers, which support tool use and task automation. Mobile devices are also important agent environments because they are widely accessible and contain users' data, sensors, and daily-use applications. Existing mobile agents mainly operate smartphones through graphical user interface (GUI) actions such as tapping, swiping, and typing, which often form long, interface-dependent sequences, cannot directly access device capabilities, and make execution boundaries difficult to define. We present \textbf{PalmClaw}, an open-source agent framework that runs natively on mobile phones and manages the sessions, memory, skills, tools, and agent loop directly on the device. PalmClaw exposes device capabilities as device tools with explicit arguments, structured results, and clearly defined execution boundaries. This design enables agents to use mobile capabilities directly while keeping each action explicit and controlled. Experiments show an 11.5\% relative improvement in task success and a 94.9\% reduction in completion time over the strongest baseline, with lower setup burden and traces illustrating how execution boundaries are applied. Code is available at https://github.com/ModalityDance/PalmClaw.
High-volume structured extraction pays a large model's latency on every item, so distilling the task into a small on-device model is attractive: comparable output at a fraction of the time and cost. We measure what that distillation actually delivers, per sub-task. Each news article is mapped to one JSON object with a short summary and five categorical labels. We distill an 8B reasoning teacher (deepseek-r1:8b) into a 0.6B student (Qwen3-0.6B; QLoRA, three seeds), and add two teacher controls: a same-size non-reasoning teacher and a larger managed pipeline. A blinded, reference-free, three-judge panel scores every arm against the full article, alongside two non-distillation baselines, few-shot prompting and constrained decoding. The student runs at about 0.8 s per article against the teacher's 39 s, and recovers 58% of the base-to-teacher gap on summary quality, beating its primary baseline (constrained decoding) by +16.8 points and few-shot prompting by a secondary +4.9. A same-size non-reasoning teacher trains a student no better than the untuned base, so the summary gain follows from the teacher's reasoning nature rather than its scale. Capabilities then split by teacher: the reasoning teacher transfers writing quality and the managed pipeline transfers label diversity, while a same-size instruction teacher's students stay more grounded on the 22 short, thin-source articles in the 93-item test set (74 versus 55 faithful), where the reasoning-lineage student fabricates. That grounding difference is a consistent ordering rather than a significant aggregate effect, and the subgroup is small, so we report it as a direction. Because no single engine wins every field, the deliverable is a per-field routing map for on-device enrichment.
Nirhoshan Sivaroopan, Albert Zomaya, Kanchana Thilakarathnacs.LG cs.AI
HAR is increasingly expected to run continuously on edge devices, yet recent LLM-based methods remain hard to deploy: raw sensor prompts are long, cloud inference adds latency and privacy risk, and fine-tuned LLM pipelines turn general-purpose models into task-specific classifiers. We present STELLA, an efficient sensor-to-LLM translation framework for on-device HAR that shifts the burden from LLM adaptation to sensor tokenization. A lightweight hierarchical tokenizer compresses an entire multi-channel inertial window into a fixed set of compact latent sensor tokens, which are projected into the embedding space of a frozen pretrained LLM and combined with a natural-language prompt for label scoring. This preserves activity-relevant temporal and cross-channel structure while keeping LLM-side computation predictable across sensor configurations. STELLA also supports on-device personalization, adapting only the lightweight tokenizer on small amounts of user-specific labelled data and augmenting inference with a local retrieval context, keeping the LLM, user data, and retrieval on device. Across seven public HAR datasets and eight benchmark settings, STELLA achieves new state-of-the-art performance, improving over prior methods by up to 11.83% F1; on-device personalization yields up to a further 21.91% F1 as user data accumulates after deployment. STELLA also outperforms representative time-series tokenizers under the same LLM pipeline and achieves real-time inference under practical mobile and edge budgets, showing that efficient sensor tokenization is a practical path toward accurate, private, and personalized LLM-based HAR on edge devices.
Maternal and newborn mortality remain among the highest in sub-Saharan Africa, where midwifery care is often delivered by nurses who lack midwifery training to international standards, and consulting authoritative guidance at the point of care is hard: the guidelines are long and connectivity is intermittent. We present MAM-AI, a medical question-answering assistant for nurse-midwives in Zanzibar that runs entirely on a commodity Android device: a question is embedded (EmbeddingGemma, 300M) and matched against a curated corpus of 87 guideline documents (63,650 passages), then answered with citations by a 4B int4 generator (Gemma 4 E4B), fully offline, with no query leaving the device. We evaluate the exact deployed configuration with a layered methodology -- retriever, generator under oracle context, end-to-end, and latency -- scored by LLM judges validated against physician rubrics. The evaluation relocates the hard problem. On-device retrieval is essentially solved: the 300M embedder ranks third of seven retrievers and rivals cloud systems, so the passages the system needs are usually found. The small generator is what remains in doubt: adding retrieved context does not improve its answers, and at 4B it cannot be both helpful and safe at once -- of two same-size candidates, the more helpful one commits genuine dangerous errors, so we deploy the other, which is about twice as faithful to its sources (as faithful as a frontier model), and recover its helpfulness with a redesigned prompt that cuts deflection from 33% to 3%. Corpus quality is decisive for the same reason: where the corpus holds the right passage the answer is specific and actionable, and where it does not it goes vague. MAM-AI is a thoroughly evaluated, open-source research prototype, not a fielded product; the system, knowledge base, benchmarks, and evaluation harness are released.
To minimize privacy concerns and inference latency on edge devices like smartphones, lightweight on-device models remain important for end-user applications. Many of these applications involve natural language classification, but deploying multiple specialized models creates a memory footprint challenge. We investigate: Can a single lightweight architecture solve multiple Speech-Adjacent (SA) classification tasks through reduction to a nuanced text similarity formulation? We propose AnySimLite, a lightweight similarity encoder that combines word-level and character-level channels. Together with a dataset transformation strategy, we evaluate AnySimLite across multiple SA classification tasks and show that it consistently achieves state-of-the-art (SOTA) or SOTA-competitive performance in few-shot settings while maintaining a low memory footprint. Even in the worst case, the performance drop remains below 7% while using $<\frac{1}{250}^{\mathrm{th}}$ of the model size of the SOTA qLLaMA_LoRA-7B baseline.
This study addresses on-device inference bottlenecks of Transformer models on Tenstorrent's Tensix architecture and proposes an operator fusion strategy that enhances data locality. RMSNorm is fused with matrix multiplication in self-attention and in the FFN, enabling back-to-back execution of memory-bound and compute-bound operators in on-chip SRAM to significantly reduce DRAM reads/writes of intermediate results and scheduling overhead. To support multi-core parallelism, a NoC-based multicast mechanism is leveraged in which row/column master nodes efficiently distribute inputs and weights across the core mesh, alleviating DRAM bandwidth contention. Experiments on the Wormhole platform with Qwen2.5-0.5B, Qwen3-0.6B, and Qwen3-4B show up to 37.44% latency reduction for attention and 15.89% for MLP, with up to 7.91% reduction per decoder layer, while Pearson Correlation Coefficient (PCC) remains above 98.75%, confirming significant end-to-end efficiency gains under numerical consistency.