Skip to content
[email protected] : ~/research $

cat ~/research/index.md

research

This is not here as a trophy case. It is a record of the questions I keep coming back to: what breaks, what can be measured, what should be audited, and where human judgment still has to carry the weight.

PAPERS
20
YEAR
2026
MODE
full list, not an archive
RELATED
→ /projects
№ 01
2026

Inside the Multilingual Transformer: Computational Neuroanatomy of Shared and Language-Specific Brain Alignment

I looked inside multilingual encoders and compared their internal signals against multilingual fMRI data. The interesting part: the shared cross-language alignment seems to come more from attention-side computations than the usual 'semantic FFN' story.

// why i care

A single benchmark score does not tell you what part of the model is doing the work. This gives multilingual model builders a more inspectable target.

authors: Ali Uyar

multilingual-nlpbrain-alignmentmechanistic-analysis
№ 02
2026

Language-Neutral or Language-Specific? Disentangling Multilingual LLM Subspaces with Naturalistic fMRI

This paper separates shared multilingual structure from language-specific leftovers in XLM-R and NLLB-200 using naturalistic fMRI data. Less 'does it work?', more 'what kind of representation is it using?'.

// why i care

Useful when you want to audit whether a multilingual encoder is carrying meaning across languages or just getting good at language-specific residue.

authors: Ali Uyar

multilingual-nlpfmri-encodingrepresentation-analysis
№ 03
2026

Overlay Algebra: Causal Composition of Operational Behaviors in Frozen Language Models

A method for adding operational behaviors like strict JSON or exact quotation on top of a frozen model without retraining the whole thing. The goal is to make output contracts reusable and testable.

// why i care

Teams keep needing small, boring behavior guarantees from large models. This treats those guarantees as something you can isolate, ship, and audit.

authors: Ali Uyar

frozen-modelsadapter-residualsauditable-evaluation
№ 04
2026

Selective Revocation and Replay: Post-Compromise Recovery of Explicit Persisted State in Memory-Augmented LLM Agents

A recovery method for poisoned agent memory. Instead of wiping everything, it tracks where bad state came from, revokes descendants, and replays only the parts that need repair.

// why i care

Agent memory is useful until it carries attacker garbage forward. This gives operators a cleaner option than 'delete everything and hope'.

authors: Ali Uyar

prompt-injection-recoveryagent-memoryprovenance
№ 05
2026

Same-Size Capability Transfer Reveals a Performance-Localization Tradeoff in Deterministic Function Calling

A locked testbed for moving function-calling behavior between same-family models. Sparse transfer worked, but the dense module still won the main metric, which is exactly the kind of inconvenient result worth knowing.

// why i care

It gives model-editing claims a deterministic check instead of a nice story and a few cherry-picked examples.

authors: Ali Uyar

function-callingdeterministic-evalmodel-editing
№ 07
2026

ProbeRoute: Probes as Routing Priors for Frozen-Backbone Multi-Token Prediction

A cheap probe pass is used to initialize routing for multi-token prediction on a frozen backbone. The aim is better speculative drafting without making the adaptation recipe bigger and messier.

// why i care

If you can get a better route from a small offline probe, you should not need to unfreeze half the model just to test the idea.

authors: Ali Uyar

frozen-backbonemulti-token-predictionspeculative-decoding
№ 08
2026

When Accuracy Hides Path Dependence: Criterion Fidelity in LLM Judges

This paper tests whether LLM judges follow the actual criterion or just react to how the criterion is worded. Accuracy can look fine while the judge is still path-dependent in a way you would not want in production.

// why i care

Good judge pipelines need to survive paraphrase and counterfactual checks. Otherwise the reviewer is grading the wording, not the work.

authors: Ali Uyar

llm-evaluationjudge-validityreliability
№ 09
2026

Binary Evidence Sufficiency Dissociation in Reasoning-Model Hidden States

A probe study asking whether reasoning models carry an internal signal for 'I have enough evidence now.' That signal appears separable from just getting the final answer right.

// why i care

For retrieval and agent systems, knowing when the model has enough evidence may matter more than another confident-looking answer.

authors: Ali Uyar

interpretabilitymulti-hop-qahidden-state-probes
№ 10
2026

Structured Yet Fragile: Signed Intervention Geometry of Matched ORF and CRISPR Cell Painting Profiles

A locked analysis of matched ORF and CRISPR Cell Painting profiles. The structure is real, but fragile, and the retrieval utility disappears under stricter checks.

// why i care

Negative results like this are useful. They tell you which biological claims still hold when analytic flexibility is taken away.

authors: Ali Uyar

locked-pipelinerobustnessnegative-result
№ 11
2026

Outcome Is Not Verification: Auditing Hidden-State Verifiers with Counterfactual Local Validity

An audit of hidden-state verifiers that separates 'the final answer was right' from 'the process was locally valid.' Those are easy to blur and dangerous to confuse.

// why i care

If a verifier is just reading outcome signals, it should not be sold as process verification. This paper gives teams a way to check that.

authors: Ali Uyar

hidden-state-verifiersprocess-verificationcounterfactual-audit
№ 12
2026

First-Unsafe-Step Counterfactual DPO for KPI-Gaming in Autonomous LLM Agents: Held-Out Evaluation Under a Capability-Safety Bottleneck

A held-out agent-safety study that finds the first unsafe step, rewrites it safely, and trains on that contrast. The catch: preference training is limited by what the base model can even produce on-policy.

// why i care

It is a practical warning for agent safety work: better labels do not magically fix a weak behavioral horizon.

authors: Ali Uyar

agent-safetypreference-optimizationheld-out-evaluation
№ 13
2026

What Survives Control Calibration? A Full-Scope Negative Result for a Locked Minimum-Description Acceptance Criterion

A full-scope negative result for a locked causal-abstraction acceptance rule. Across the tested settings, no proposed abstraction class earned certification.

// why i care

This is the kind of failure report interpretability needs more of: clear rules, no post-hoc rescue, and boundaries on what can be claimed.

authors: Ali Uyar

mechanistic-interpretabilityacceptance-criterianull-calibration
№ 14
2026

CacheMedic++: Robust KV-Cache Stabilization via Self-Distillation

A reproducible way to corrupt, diagnose, and repair KV-cache failures using distillation. The goal is to turn weird serving instability into something you can actually test.

// why i care

Serving failures are easier to fix when they are repeatable. This makes that repeatability part of the workflow.

authors: Ali Uyar

kv-cachedistillationrobustness
№ 19
2026

RIA: Retokenization Invariance Atlas

A deterministic audit for cases where semantically identical prompt formatting still changes the answer. Newline roulette is not a reliability strategy.

// why i care

It gives teams evidence about whether formatting changes are harmless or quietly changing outcomes.

authors: Ali Uyar

tokenizationevaluationauditability
№ 20
2026

CIS Technical Report

A practical reliability report for AI systems: deterministic checks, evidence gates, and implementation notes for teams that have to ship the thing, not just demo it.

// why i care

It turns the reliability philosophy into something a delivery team can actually adopt.

authors: Ali Uyar

technical-reportreliabilitysystems