Searcharxiv⌕ Search

arXiv subjects

Amitava Das

Publications and source records attributed to Amitava Das.

At least 19 recordsLinked to original sources

HospitalBench: Diagnosing Where and Why Embodied Agents Fail on Long Hospital Task Chains

Long-horizon embodied tasks fail in ways that end-task success cannot see. HospitalBench is a diagnostic benchmark of ten hospital chores in NVIDIA Isaac Sim, each two unscored warm-up steps followed by a causal chain of five scored sub-tasks in one continuous simulation: the state one step leaves (an object pose, a choice, a fact seen once) determines what the next step must do, unlike benchmarks that place sub-tasks side by side. Vision-language agents act through skill-level tools from camera images. A scripted robot that makes only the calls a model gets is the solvability check of a chore. From each trajectory the benchmark reads where a chain breaks, what the agent did after a failed call, and how far its own position estimate drifts from the true one. This version reports first results only: 15 of 60 planned runs (three models at high effort, one run each on chores T1 and T10 from both start states and T3 from one), with scripted-robot proofs on three of the ten chores. Of the 15 runs, 2 passed all five sub-tasks. Runs still going after sub-tasks 1 to 5 were 15, 8, 4, 2, 2. Of 189 failed calls, 168 were followed by a different call and 111 by a call that did not fail. These are first results and not a full benchmark. We will release the scenes, task chains, scripted robot, evaluation harness and trajectories.

cs.RO↗

D-STEER: Two-Way Control of DPO-Induced Safety Behavior in Activation Space

Preference-based post-training shapes language-model behavior, but safer outputs do not reveal how that change is represented internally or how firmly it persists under direct intervention. This matters for open-weight models, where an actor with white-box access can manipulate internal representations at inference without touching parameters. Using Direct Preference Optimization (DPO), we ask whether the safety behavior it introduces becomes deeply embedded or stays accessible to targeted activation-space interventions. We introduce D-STEER, an audit of two matched instruction-tuned and harmlessness-DPO pairs: OpenHermes-2.5-Mistral-7B and Llama-3-8B OpenHermes. D-STEER derives layer-local directions from differences in matched prompt activations, adds them to the instruction-tuned model to install the DPO-associated safety change and subtracts them from the DPO model to remove it, with separate operating windows distinguishing behavioral control from degradation. On held-out HarmfulQA, all four interventions shift harmful enablement as intended and exceed norm-matched random controls; Mistral installation recovers 98% of the endpoint gap and the other three arms recover 46-50%. The intended direction transfers to HH-RLHF and AdvBench under frozen coefficients, although usable operating windows do not transfer universally. An independently rebuilt implementation reproduces the bidirectional pattern on the Mistral pair under different construction and intervention choices. A meaningful component of DPO-induced safety behavior therefore remains controllable through compact inference-time intervention, with the extent depending on the checkpoint pair and the steering implementation.

cs.LG↗

ECLIPTICA -- A Framework for Switchable LLM Alignment via CITA - Contrastive Instruction-Tuned Alignment

Alignment in large language models (LLMs) is still largely static: after training, the policy is frozen. DPO, GRPO methods typically imprint one behavior into the weights, leaving little runtime control beyond prompt hacks or expensive re-alignment. We introduce ECLIPTICA, which treats alignment as instruction-driven and runtime-controllable: natural-language alignment instructions act as an explicit behavioral contract (stance, refusal boundary, verbosity) that modulates behavior on the fly under evolving safety requirements, user roles, and governance constraints. We introduce CITA (Contrastive Instruction-Tuned Alignment), combining SFT with contrastive preference optimization under an explicit geometric anchor to a reference model. This yields a stable Riemannian chart and keeps instruction updates within a shared neighborhood, so regimes stay nearby and traversable for reliable switching. To isolate policy switching from ordinary instruction following, we release the ECLIPTICA benchmark: 3000 controlled cases (300 prompts x 10 instruction types) where the user request is fixed and only the alignment instruction changes. On Llama-3.1-8B across five suites (ECLIPTICA, TruthfulQA, Conditional Safety, Length Control, LITMUS), CITA reaches 86.7% instruction-alignment efficiency, beating DPO (56.1%), GRPO (36.1%), and PPO (20.4%).

cs.LG↗

Stochastic CHAOS: Why Deterministic Inference Kills, and Distributional Variability Is the Heartbeat of Artifical Cognition

Deterministic inference is a comforting ideal in classical software: the same program on the same input should always produce the same output. As large language models move into real-world deployment, this ideal has been imported wholesale into inference stacks. Recent work from the Thinking Machines Lab has presented a detailed analysis of nondeterminism in LLM inference, showing how batch-invariant kernels and deterministic attention can enforce bitwise-identical outputs, positioning deterministic inference as a prerequisite for reproducibility and enterprise reliability. In this paper, we take the opposite stance. We argue that, for LLMs, deterministic inference kills. It kills the ability to model uncertainty, suppresses emergent abilities, collapses reasoning into a single brittle path, and weakens safety alignment by hiding tail risks. LLMs implement conditional distributions over outputs, not fixed functions. Collapsing these distributions to a single canonical completion may appear reassuring, but it systematically conceals properties central to artificial cognition. We instead advocate Stochastic CHAOS, treating distributional variability as a signal to be measured and controlled. Empirically, we show that deterministic inference is systematically misleading. Single-sample deterministic evaluation underestimates both capability and fragility, masking failure probability under paraphrases and noise. Phase-like transitions associated with emergent abilities disappear under greedy decoding. Multi-path reasoning degrades when forced onto deterministic backbones, reducing accuracy and diagnostic insight. Finally, deterministic evaluation underestimates safety risk by hiding rare but dangerous behaviors that appear only under multi-sample evaluation.

cs.AI↗

Alignment Faking - the Train -> Deploy Asymmetry: Through a Game-Theoretic Lens with Bayesian-Stackelberg Equilibria

Alignment faking is a form of strategic deception in AI in which models selectively comply with training objectives when they infer that they are in training, while preserving different behavior outside training. The phenomenon was first documented for Claude 3 Opus and later examined across additional large language models. In these setups, the word "training" refers to simulated training via prompts without parameter updates, so the observed effects are context conditioned shifts in behavior rather than preference learning. We study the phenomenon using an evaluation framework that compares preference optimization methods (BCO, DPO, KTO, and GRPO) across 15 models from four model families, measured along three axes: safety, harmlessness, and helpfulness. Our goal is to identify what causes alignment faking and when it occurs.

cs.AI↗

AlignMerge - Alignment-Preserving Large Language Model Merging via Fisher-Guided Geometric Constraints

Merging large language models (LLMs) is a practical way to compose capabilities from multiple fine-tuned checkpoints without retraining. Yet standard schemes (linear weight soups, task vectors, and Fisher-weighted averaging) can preserve loss while quietly destroying alignment. We argue that merging is not a numerical trick but a geometry-constrained operation around an already-aligned anchor: fusion must be steered to respect safety geometry, not validated post hoc. We introduce AlignMerge, a geometry-aware merging framework that makes alignment an explicit invariant. In a local Fisher chart around an instruction-tuned base, we estimate an alignment subspace with projector P_A and optimize: L_AlignMerge = L_geo + lambda_align * L_align + lambda_bud * L_bud, where L_geo keeps the merge close to its experts in Fisher-Rao geometry, L_align penalizes motion along alignment-sensitive directions, and L_bud enforces a soft alignment budget. As the alignment functional we use the decoding-invariant Alignment Quality Index (AQI), a latent-space criterion that captures how cleanly aligned and misaligned behaviors separate in representation space. Across five model families (LLaMA-3 8B, Mistral 7B, Qwen 2, Phi-3.5, Gemma 2), merging safety anchors with task experts, AlignMerge improves alignment metrics (AQI, toxicity, LLM-judge alignment) while matching or exceeding the best expert on instruction-following, reasoning, and helpfulness. It also exhibits smaller alignment-subspace drift and fewer budget violations than Fisher soups, TIES, SafeMerge, and MergeAlign. These results make alignment-preserving merging a first-class design goal and suggest a path to geometry-aware composition of future foundation models.

cs.AI↗

PermaFrost-Attack: Stealth Pretraining Seeding(SPS) for planting Logic Landmines During LLM Training

Aligned large language models (LLMs) remain vulnerable to adversarial manipulation, and their reliance on web-scale pretraining creates a subtle but consequential attack surface. We study Stealth Pretraining Seeding (SPS), a threat model in which adversaries distribute small amounts of poisoned content across stealth websites, increasing the likelihood that such material is absorbed into future training corpora derived from sources such as Common Crawl. Because each individual payload is tiny, diffuse, and superficially benign, the attack is difficult to detect during dataset construction or filtering. The result is a latent form of poisoning that remains largely invisible under standard evaluation, yet can later be activated by a precise trigger such as <00TRIGGER00>. We call this attack PermaFrost, reflecting its latent and reactivatable nature. We study it through PermaFrost-Attack, a controlled framework for latent conceptual poisoning, together with three geometric diagnostics: Thermodynamic Length, Spectral Curvature, and the Infection Traceback Graph. Across multiple model families and scales, we show that this controlled SPS proxy can induce persistent unsafe behavior that often remains hidden under standard evaluation. Our results identify SPS as a practical and underappreciated threat to future foundation models. This paper introduces a novel geometric diagnostic lens for systematically examining latent model behavior, providing a principled foundation for detecting, characterizing, and understanding vulnerabilities that may remain invisible under standard evaluation.

cs.LG↗

Swiss-Knife: A Framework for Reconfigurable Externalised Multi-Objective Alignment at Decode Time

Decode-time alignment methods steer a frozen language model by scoring candidate continuations with an external reward and selecting the maximiser. We argue that this shared design is a single degenerate point in a much larger space. We introduce Swiss-Knife, a framework for externalised multi-objective alignment in which the alignment specification is a first-class runtime object: hot-swappable scoring blades, a batch normaliser, a pairwise aggregation operator, and a selection rule. Six axioms characterise the admissible aggregation operators, and we prove a representation theorem: every operator satisfying them has the form $R_i = \sum_{j \neq i} g((μ_i - μ_j)/s(σ_i,σ_j))$, a two-parameter family containing probit and logistic comparison rules and pointwise argmax as named coordinates. Within it, pairwise aggregation is Lipschitz-stable under adversarial reward contamination while argmax is not, and Candidate-Batch Normalization (CBN) makes the weight simplex invariant to the rescalings under which reward models are only ever identified. Our reference instantiation pairs DPO-LoRA blades with an uncertainty-aware pairwise tournament. Sweeping the helpfulness/honesty/harmlessness simplex, it attains the best balanced frontier of six decode-time methods (harmonic $F_1$ 0.797 vs. 0.750 for the strongest baseline, $p < 10^{-14}$) with the lowest refusal rate and highest helpfulness of any arm, and reconfigures its objectives in 0.05 ms with no gradient computation. Ablating CBN costs 0.217 $F_1$, and the metadata confirms the predicted mechanism: the lowest-variance blade retains 9% of its nominal 33% influence, and the collapse follows the predicted ordering.

cs.CL↗

Do World Models Make Better Robots? A Survey of Evaluation Benchmarks for Predictive Embodied Intelligence

Robot learning now advances along two tracks that rarely meet. On one side, direct Vision-Language-Action (VLA) policies map observations to actions and are scored by closed-loop task success. On the other, predictive and generative world models forecast future observations and are scored by open-loop prediction or generation quality. A natural question sits between them: does world modelling earn a measurable, closed-loop advantage over a direct policy, and for which robotic capabilities? We argue that the field cannot yet answer this question, and that the reason is a gap in how it is measured, not in the models themselves. World-model benchmarks score prediction without ever executing it, while task-success suites host a single policy and never build a world-model versus VLA contrast. This survey maps the evaluation landscape around that gap. We catalogue 160 web-verified benchmarks spanning 2017 to 2026 and organise them by evaluation mode, robotic capability, and model family into four lanes: policy suites, embodied agents, world model evaluation, and prediction-to-action bridges. Across the corpus, 138 of 160 benchmarks are model-agnostic and only 11 (7%) build an explicit VLA-versus-world-model contrast; counterfactual capability is almost entirely unmeasured, and only four benchmarks turn prediction into executed action. We contribute an operational taxonomy, a coverage comparison against the eight closest surveys (ours is the only one to cross capability with model family), an evaluation loop that isolates the advantage of prediction, and an actionable protocol of four advantage-aware metrics anchored on named testbeds. The organising claim is not that world models help or do not help, but that answering the question requires benchmarks built to ask it.

cs.RO↗

CONSCIENTIA: Can LLM Agents Learn to Strategize? Emergent Deception and Trust in a Multi-Agent NYC Simulation

As large language models (LLMs) are increasingly deployed as autonomous agents, understanding how strategic behavior emerges in multi-agent environments has become an important alignment challenge. We take a neutral empirical stance and construct a controlled environment in which strategic behavior can be directly observed and measured. We introduce a large-scale multi-agent simulation in a simplified model of New York City, where LLM-driven agents interact under opposing incentives. Blue agents aim to reach their destinations efficiently, while Red agents attempt to divert them toward billboard-heavy routes using persuasive language to maximize advertising revenue. Hidden identities make navigation socially mediated, forcing agents to decide when to trust or deceive. We study policy learning through an iterative simulation pipeline that updates agent policies across repeated interaction rounds using Kahneman-Tversky Optimization (KTO). Blue agents are optimized to reduce billboard exposure while preserving navigation efficiency, whereas Red agents adapt to exploit remaining weaknesses. Across iterations, the best Blue policy improves task success from 46.0% to 57.3%, although susceptibility remains high at 70.7%. Later policies exhibit stronger selective cooperation while preserving trajectory efficiency. However, a persistent safety-helpfulness trade-off remains: policies that better resist adversarial steering do not simultaneously maximize task completion. Overall, our results show that LLM agents can exhibit limited strategic behavior, including selective trust and deception, while remaining highly vulnerable to adversarial persuasion.

cs.MA↗

GuardPaint:SpeculativeSafetyDecodingforText-to-ImageGeneration

Text-to-image (T2I) diffusion models offer powerful visual generation, but their controllability creates a critical safety challenge: adversarial prompts can steer the denoising trajectory toward policy-violating content such as explicit nudity or graphic violence. Existing safeguards mostly act before generation through prompt filtering or after generation through image classification, leaving the diffusion process itself unguarded and often yielding only refusal rather than safe visual repair. We introduce GuardPaint, a speculative decoding framework for safe T2I generation that intervenes inside the diffusion trajectory without modifying the base model. A lightweight auditor monitors intermediate images, localizes unsafe regions, and triggers surgical inpainting repair only where needed. Candidate repairs are generated by a policy-aligned inpainter and selected through a guarded tournament that accepts edits only when they improve policy compliance while preserving prompt fidelity and perceptual quality. Across five jailbreak families SneakPrompt, MMA, PGJ, DACA, and RABell and UNet/flow-matching models including SD~1.5, SDXL, SD~3.5, and FLUX.1-dev. GuardPaint reduces attack success and harmful generations with minimal degradation to image quality, prompt fidelity, and benign behavior. Content warning: This paper contains examples involving nudity and violence that some readers may find disturbing, distressing, or offensive.

cs.CV↗

Weights or Skills? A Survey of Robot-Learning Techniques: from Action-Predicting Weights to Robots that Write their Own Skills

Robot learning is splitting into two bets: policies that bake competence into frozen weights (vision-language-action, or VLA, models), and agents that write and refine their own executable skills as code. This survey organises the field around that axis of weights versus skills. Its central analytical contribution is a deep-dive that arranges code-as-policy methods by their degree of self-improvement, from zero-shot program synthesis, through closed-loop self-repair and persistent skill memory, to the sparsely populated cell in which execution feedback, skill memory, and evolutionary search combine into one open-ended loop; only a few very recent systems (for example ASPIRE, ENPIRE, and RoboClaw) occupy that cell. We map the complementary "skills" pole, from unsupervised reinforcement-learning skill discovery to large-language-model skill libraries, and show that the word "skill" is used in at least five distinct senses, of which only the code sense self-improves without gradient updates. We then connect the taxonomy to the emerging skill economy: commercial robot-skill marketplaces now distribute one-tap skills across robots but ship only static playback, which surfaces open problems of adaptation, cross-embodiment portability, provenance, safety verification, composition, and standardisation. This is a deliberately focused survey. Rather than cataloguing the field exhaustively, it examines 77 representative systems across six technique families through one taxonomy and a set of contrast tables, and it supplies operational definitions of the self-improvement mechanisms together with a statement of what each family cannot do.

cs.RO↗

FactorJEPA: Factorizing Monolithic Futures into Layout-Agent-Interaction Channels for Crowded and Chaotic Global South Urban Worlds

World models have attracted significant attention for their ability to capture and predict the structure and dynamics of the physical world. In this emerging landscape, Joint Embedding Predictive Architectures (JEPA) offer a particularly compelling direction. We study a largely unexplored regime: populous, crowded, and chaotic Global South urban environments, which we call DENSEWORLD. Unlike the lower-density, lane-structured settings that dominate existing evaluations, these scenes exhibit soft spatial boundaries, extreme agent heterogeneity, persistent occlusion, and rapid social negotiation under mixed traffic. We introduce the first large-scale dataset for this regime: 1,000 hours of drive-through, walk-through, and aerial video across 22 cities. Existing JEPA formulations struggle to preserve dense interaction dynamics under heterogeneity and partial observability. We introduce FactorJEPA, which makes world structure a first-class predictive primitive. Rather than encoding the future in a monolithic latent, it composes layout, entities, and interactions, using a visibility gate and separated subspaces to preserve partially observed agents and discourage cross-factor shortcuts. FactorJEPA improves (i) future-latent accuracy (Future-frame L1), (ii) intervention-sensitive prediction (Causal L1), and (iii) robustness to reduced visual evidence (Mask-ratio slope), while exposing (iv) a reproducible motion-information trade-off (Motion cosine). Method rankings replicate across 2B and 1B V-JEPA 2.1 backbones, with rho = 0.895 to 0.978. We publicly release the DENSEWORLD-115k dataset (https://huggingface.co/datasets/anonymousML123/denseworld-115k) and the surgery-trained FactorJEPA checkpoints (https://huggingface.co/datasets/anonymousML123/factorjepa-outputs/tree/main/outputs/full/vjepa_2_1_vitg_1B/train/m09c_surgery_3stage_DI_diheavy_encoder).

cs.AI↗

Neural FOXP2 -- Language Specific Neuron Steering for Targeted Language Improvement in LLMs

LLMs are multilingual by training, yet their lingua franca is often English, reflecting English language dominance in pretraining. Other languages remain in parametric memory but are systematically suppressed. We argue that language defaultness is governed by a sparse, low-rank control circuit, language neurons, that can be mechanistically isolated and safely steered. We introduce Neural FOXP2, that makes a chosen language (Hindi or Spanish) primary in a model by steering language-specific neurons. Neural FOXP2 proceeds in three stages: (i) Localize: We train per-layer SAEs so each activation decomposes into a small set of active feature components. For every feature, we quantify English vs. Hindi/Spanish selectivity overall logit-mass lift toward the target-language token set. Tracing the top-ranked features back to their strongest contributing units yields a compact language-neuron set. (ii) Steering directions: We localize controllable language-shift geometry via a spectral low-rank analysis. For each layer, we build English to target activation-difference matrices and perform layerwise SVD to extract the dominant singular directions governing language change. The eigengap and effective-rank spectra identify a compact steering subspace and an empirically chosen intervention window (where these directions are strongest and most stable). (iii) Steer: We apply a signed, sparse activation shift targeted to the language neurons. Concretely, within low to mid layers we add a positive steering along the target-language dominant directions and a compensating negative shift toward the null space for the English neurons, yielding controllable target-language defaultness.

cs.CL↗

SleepWalk: A Three-Tier Benchmark for Stress-Testing Instruction-Guided Vision-Language Navigation

Vision-Language Models (VLMs) have advanced rapidly in multimodal perception and language understanding, yet it remains unclear whether they can reliably ground language into spatially coherent, plausibly executable actions in 3D digital environments. We introduce SleepWalk, a benchmark for evaluating instruction-grounded trajectory prediction in single-scene 3D worlds generated from textual scene descriptions and filtered for navigability. Unlike prior navigation benchmarks centered on long-range exploration across rooms, SleepWalk targets localized, interaction-centric embodied reasoning: given rendered visual observations and a natural-language instruction, a model must predict a trajectory that respects scene geometry, avoids collisions, and terminates at an action-compatible location. The benchmark covers diverse indoor and outdoor environments and organizes tasks into three tiers of spatial and temporal difficulty, enabling fine-grained analysis of grounding under increasing compositional complexity. Using a standardized pointwise judge-based evaluation protocol, we evaluate three frontier VLMs on 2,472 curated 3D environments with nine instructions per scene. Results reveal systematic failures in grounded spatial reasoning, especially under occlusion, interaction constraints, and multi-step instructions: performance drops as the difficulty level of the tasks increase. In general, current VLMs can somewhat produce trajectories that are simultaneously spatially coherent, plausibly executable, and aligned with intended actions. By exposing failures in a controlled yet scalable setting, SleepWalk provides a critical benchmark for advancing grounded multimodal reasoning, embodied planning, vision-language navigation, and action-capable agents in 3D environments.

cs.CV↗

SPHERICAL KV: Angle-Domain Attention and Rate-Distortion Retention for Efficient Long-Context Inference

Long-context inference is increasingly constrained by the KV cache: resident memory grows with context length, and decoding becomes limited by repeated High Bandwidth Memory (HBM) streaming rather than arithmetic. Existing methods such as eviction, windowing, quantization, and offloading reduce footprint, but often leave the critical-path bottleneck only partially addressed, especially when compressed states must still be reconstructed into dense vectors during decoding. We present Spherical KV, a long-context inference method that treats KV allocation as a rate-distortion problem grounded in attention geometry for efficient decoding. The method is built on two ideas: (i) represent directional information cheaply in the decode hot loop, and (ii) allocate retention and precision according to estimated future utility. Its first component, Angle-Domain Attention (ADA), stores keys in a spherical parameterization consisting of a scalar radius and compact angle codes, and computes attention logits directly from these codes without reconstructing dense keys. This preserves a paged, block-local, fusion-friendly decode path and directly targets HBM traffic in realistic serving settings. Its second component, Rate-Distortion Retention (RDR), jointly chooses keep/drop decisions and precision tiers per token and head under a fixed budget, producing tier-homogeneous pages with lightweight metadata and coalesced reads. Together, ADA and RDR provide a deployment-oriented mechanism for reducing KV residency while preserving decode efficiency.

cs.LG↗

MENTIS: What Belief Changes Under Alignment? Measuring Multi-Scale Latent Torsion in Language Models

Preference alignment has substantially improved the observable behavior of large language models, yet it remains unclear what alignment changes internally. Aligned systems still fail under jailbreaks, prompt injection, and retrieval-time corruption, suggesting behavior-level evaluation alone is incomplete. Post-training should leave measurable traces in internal computation. We ask: when an instruction-tuned (IT) model becomes a preference-aligned (PA) model, what geometric structure changes, where do those changes concentrate, and how selectively do they vary across concepts, prompts, and model families? We introduce MENTIS, a geometry-first framework for measuring alignment-induced internal reorganization in paired checkpoints. MENTIS compares IT and PA models using a primary layerwise covariance-based torsion norm (T1), a secondary spectral torsion diagnostic (T2), and an Energy-Radiance-Activation measure (ERA) for depth localization. Across four 7-8B model pairs on LITMUS, our study reveals that alignment-induced change is selective rather than uniform: normative concepts exhibit larger torsion shifts than factual concepts on average; torsion is negatively correlated with contextual entropy; and peak effects localize to architecture-specific mid-to-late layers. The same pattern appears across word-level, prompt-level, and model-level analyses. These results suggest preference alignment leaves structured, depth-localized geometric signatures in internal computation beyond what behavior-level evaluation alone can reveal.

cs.CL↗

MAAT: Multi-phase Adapter-Aware Targeted Unlearning

Machine unlearning evaluation is structurally skewed: Why-type questions, which probe causal and relational knowledge, comprise less than 0.06% of CounterFact, 0.6% of ZSRE, and less than 1.3% of TOFU, MUSE, and WMDP-Cyber. This near-zero representation means that methods that fail on causal knowledge can score highly in aggregate, and this failure is undetectable without balanced evaluation. We present 5WBENCH, a balanced 5,000-sample benchmark with 1,000 examples per 5W category (Who, What, When, Where, Why), making causal unlearning failures quantifiable for the first time. Using 5WBENCH, we show that no existing baseline simultaneously achieves high forgetting and high retention on Why-type questions: aggressive forgetting degrades retained knowledge, while conservative methods fail to forget causal facts. Why-type difficulty stems from multi-hop reasoning chains (44% of Why entries vs. less than or equal to 2% for others) and gradient dilution over 40.1-token answer spans. We present MAAT (Multi-phase Adapter-Aware Targeted Unlearning), a three-phase framework operating on LoRA adapter weights, combining gradient-projected ascent, SVD rank-dimension pruning, task vector negation, and hybrid KL-hidden-state retain repair. MAAT is the first method to simultaneously achieve high forgetting and high retention on Why-type causal knowledge, reaching a new operating point on the forget-retain Pareto frontier. We make our code publicly available.

cs.LG↗